Extraction of Blood Cell Image Classification using Convolution Neural Network

This document discusses the concept of blood cell image classification using convolution neural network and transfer learning pre-trained model. It explains the process of blood cell extraction and classification using CNN and highlights its importance in medical diagnosis.

Extraction of blood cell image classification using convolution neural network
and transfer learning pre-trained model by using python
1

ABSTRACT
It has summarized about the concept of blood cell extraction which become easier to divide into
different multilayer cells. It has been identified that acquisition of image that could be arranged
in short terms and also wide spread stress fibres as long bundles. Convolutional neural network
technique, which composed by pooling layers that support in blood cell classification. The
primary research project is to generate the specific features from image processing. In order to
identify the extraction of blood cell image the report provide the learning about the hierarchical
feature, means that computed by multi layers. It consists of large number of data sets but it easily
classified with the help of convolutional neural network. The concept related convolutional
neural network because it help for identifying structure, colours, texture of blood cell image. In
term of medical, it is the most efficient method extraction of blood cell image classifications.
2

Contents
INTRODUCTION...........................................................................................................................3
LITERATURE REVIEW................................................................................................................9
RESEARCH METHODLOGY.....................................................................................................15
DATA INTERPRETATION.........................................................................................................21
RESULT AND FINDING.............................................................................................................37
CONCLUSION..............................................................................................................................39
RECOMMEDNDATION..............................................................................................................43
REFLECTCTION..........................................................................................................................45
REFERENCES..............................................................................................................................50
3

Title: Extraction of blood cell image classification through convolution neural network and
transfer learning (pre-trained) model by using python.
INTRODUCTION
The extraction of blood cell image classification can be possible with the help of
convolutional neural network. it can be defined as manipulating an image to achieve standards
that categorized into different images. It is well known method in term of blood cells
classification which may include red blood, white blood and platelets (Sahlol, Kollmannsberger
and Ewees, 2020). During blood extraction, leucocyte plays important role in term of human
immune functionality. Generally, it mainly identified the hematologists that can use for purpose
of granulated information and size or shape in leukocytes. It can be divided into different blood
cells such as basophil, eosinophil and non-granular cells. It often uses the basic information on
the basis of categorizes therefore, research study of blood cell classification which has important
in term of medical diagnosis.
The microscopic blood image of every patient play important role for purpose of diagnosis
and also managed with many diseases. In blood cells, there are various type of cells such as
white blood cells and red blood cells. This type of blood cell play important role in human body
which maintain immune system and fight against with viruses, bacteria and microbes. These are
extremely 5000 to 10000 WBC in adult’s human body, having values outside in particular
specified manner. It has increased to multiple diseases such as anemia, deficiency disease. That’s
why, primary reason for using convolutional neural network for blood cell image classification.
In this way, it can easily identify any problem that occurs in the human body (Coulibaly, Kamsu-
Foguem, Kamissoko and Traore, 2019). It is consider as modern technology that always support
for collecting large information about blood cells. Nowadays, the study was conducted for the
extraction of blood cell image classification. In this study, it mainly focused on the convolutional
neural network and use as a model where it will create deep learning architecture. It also used in
the different classification process of generate blood cell images. Main feature is that when it
automatically extracting large amount of cells and successful provided through architecture. As a
result, it is obtaining through multiple classifier on the blood cell architecture and also compare
with other one. Sometimes, it will identify the best success rate which mainly obtained by
Quadratic discriminant analysis. On the other hand, various type of classifier along with
4

extracted attributes and also increase success rate. WBC microscopic blood image accessible
from the convolutional neural network. In order to adopt and compare with other one.
A convolutional neural network is based on technique that help for identifying data in the
form of images. It can take action as input while assign into different aspects that able to
differentiate one from another (Coulibaly, Kamsu-Foguem, Kamissoko and Traore, 2019). The
process must be used the pre-processing which require for filtering the characteristic so that it
become easier to identify in proper manner. The entire architecture of CNN is analogous that
establish the connectivity in pattern of neuron in human brain. Recently, Convolutional neural
network has been widely implemented in terms of image classification (Jokhio and Jokhio,
2019). In order to achieve the unprecedented result or outcome. In this way, it is significantly
improving the overall performance of different medical imaging medium or applications. CNN is
highly efficient in terms of image processing, since features may be used in different images and
generating the output. It also demonstrates the optimization process which consider specific
parameters. Therefore, it has performed to minimize the difference between ground label in
blood images and generating results.
It is the best way to show the clearance in the performance of cell image while it can be
classified into proper manner. Apart from that Transfer learning mainly refers to the migration of
training model which consists of multiple parameter in the model. In this way, it helps for train
new model while identifying different layers in the image classification (Umapathy and et.al.,
2019). It is primarily used as base line of network that contain specific layer with randomized
determine the weights. It became useful for facilitating the transmission features in the same
layer as Xception which transferred in the similar model. In additionally, Xception is basically
used to categories the RGB channel which may consider as input and take single channel map.
Multiple layers in cell image classification, they are connected with each other and create as
single data set. Afterwards, it can be used the original learning rate to easily train entire layer of
blood cell.
In research study, it has proposed the Convolutional Neural Network method that will
help for classifying the blood cells and shown in the form of images. It has recognized the
classification of cells with rate of 94.2%. it will be used 100 images to identify their unique
properties. Thus, it implements the automatic threshold and adaptive contour for segment cells. It
helpful for identifying the errors if in case it doesn’t found the extraction in blood cells. The
5

recognize rate of blood cell is becoming higher because of Convolutional neural network which
classified into different manner (Coulibaly, Kamsu-Foguem, Kamissoko and Traore, 2019).
Therefore, it is implementing an automated and high throughput blood cell classification method,
it could be enabling technology to improve overall medical diagnosis capabilities, predicts actual
outcome and result. The primary objective of research project is to use CNN method for
classifying the blood cell image.
The report will discuss about the effectiveness of convolution neural network and transfer
learning model that easily monitored the activities within blood cells. The technique will help for
recording large number of data which called pre-processor data set. Sometimes, it automatically
turns into the 12,444 blood cells and enhanced towards image classification. With the
improvement in the overall performance of computer through convolutional neural network,
which has become most popular technique that record image of classification. Here, it become
easier for researcher to understand structure (Qiu and et.al., 2019).
In this report, it will propose the architecture that mainly combined with CNN
(convolutional neural network). It also integrated with the features from CNN and perform the
activities of blood cell classification (Coulibaly, Kamsu-Foguem, Kamissoko and Traore, 2019).
On the other hand, CNN must be used the pre-trained model during dataset of image collection
and retain its overall parameters. The pre-trained CNN model is mainly used for purpose of pre-
processed training which considered as input that extract and save obtained features
(Loganathan, 2019). The CNN model can perform the activities and apply multiple size and
weight matrix. In order to generate various features maps in processing. The proposed blood cell
classified by technique or method which contribution towards result. The research study will
identify the advantage of convolutional neural network and transfer learning in the extraction of
blood cell image classification by using python programming.
Background study
Researches about the extraction of blood cell classification in recent years, based on
Convolutional neural network technique. it is the most common approach that applicable image
processing algorithm for consists of different steps such as morphological, clustering,
segmentation, extraction, evaluation and classification. It also adopted the machine learning
technique in order to detect and classify blood cell images. Furthermore, Convolutional statistical
feature such as correlation that are extracted for give input in machine learning. In mostly, it
6

should be needed to decide about the features that must be utilised in order to increase blood cell
image classification accuracy. For instances, Transfer learning will help for extract high level
features which are automatically perform at same time. Therefore, it has proposed the
convolutional neural network architecture to determine abnormal and normal blood cell images.
In past decade, CNN is considered as technique which help for reducing process time and allow
to skip different steps. In this way, it can easily identify better result or outcome.
The segmentation process can help for detecting the blood cell classification image while
identifying background, plasm of blood image by using image processing (Coulibaly, Kamsu-
Foguem, Kamissoko and Traore, 2019). It is based on size, shape, colour and segmentation. It
has been proposed and combined with other technique for detecting and segmenting. Extraction
features representing as important step to classify the area, compactness, multiple layer and
distribution. Through existing theories, it works have done and also classified in context of
feature extraction. Furthermore, evaluation process can be implemented during classification
which help for generating numeric metric such as accuracy. It is the most important concept for
identifying overall efficiency and performance. The prediction may be classified into positive
and negative ways. During convolution neural network technique, it should be considered
different parameters that mainly define as testing and training protocols.
In the process of implementing the machine learning technique in term of blood cell
image classification, identifying meaning features that representation lies at core of success rate
into generate desirable outcome. The Image analysis by using computer based analysis and also
used the software technique with hand-engineered features (Sahlol, Kollmannsberger and Ewees,
2020). Sometimes, it also supported in the decision making where in case if it has found any
problem within blood cells. In past decade, demand of process has been identified where
expertise in analysing overall angle, size, position and back ground of image. In order to resolve
the challenges by devising hand-engineered features that easily capture variation into specified
manner.
Deep learning model is primarily used to cascade of multi layers, representing the raw
information or data. It applied high level features that supported to abstract from lower layers
and also helping for decision making. As a resulting, it acquire end to end features and classified
blood cell image in different manner. During image recognition, it is an essential source of
various data which mainly lies in the relationship among pixels. CNN, it would be created the
7

classes of model and also designed to exploit distribution information by using receptive fields,
pooling layer and shared weights. The ImageNet architecture is representing the visual image
and easily boosted the overall performance of convolutional neural network. In another way, it
will be classified natural image of blood cells in proper manner. There are several CNN such as
ResNet, VDDNet and GoogLenet that are demonstrated significantly improvement in the
challenge and issue. Another important model such as Xception, which was proposed that help
for depth analysis. In order to identify separable convolutions. This type of model is applicable
in the extraction of blood cell classification because it can outperform inception on the imageNet
during data classification (Sahlol, Kollmannsberger and Ewees, 2020). Convolutional neural
network has been categorised into variant which are densely connected with network system. It
also utilised the entire network architecture which are supposed to establish connection between
different layers. The imageNet model has achieved the improvement plan over start of art while
enhancing the computation action through different parameters. It also promising the overall
performance of convolutional neural network which accompanied after checking availability of
large amount of data or information. Sometimes, it has increased uncertainty due to medical
imagery transfer learning.
On the other hand, it applicable pre-trained model which ether find tuned on the specific
data, extracted to recognition task (Sahlol, Kollmannsberger and Ewees, 2020). The model has
gained the knowledge about the entire functionality of CNN and then afterwards, it can examine
the generic features from large datasets such as underlying operation through ImageNet model.
In previously, it has learned the skill to improve situation or condition in generalised manner. In
this way, it has recognised the CNN trained on the dataset that could be served as extraction
feature during blood cell classification through computer vision tasks. It also improved the
overall performance and also compare with different methods.
Transfer learning
It is the most popular method or technique in term of computer vision because it allows
us to build the accurate models. By using transfer learning, it is mainly starting with learning
process that automatically scratch and start from different patterns. It have been learned when
resolving the problem and issue. It became leverage previous learning about convolutional neural
network for calculating the different multilayers of blood cell images. Sometimes, it also
undertaken as deep learning version of expression (Srivastava and et.al., 2019). During computer
8

vision, transfer learning is expressed through use of pre-trained models. This type of model was
trained on the large benchmark dataset to resolve problem in proper manner. Accordingly, it can
be used the computational cost of training such as models which become common practices to
import or export. A comprehensive review the overall extraction of blood cell image and identify
the entire performance of trained models.
Convolutional neural network
It is based on the pre-trained model which mainly used as transfer learning on large scale.
In particularly, CNN has been shown into excel format in wide range of computer vision. It has
performed the different task to identify accurate result or outcome (Islam, Mubassira and Das,
2019). It is higher performance and easiness during training. In another way, it is to be
considered the main factor which driving the popularity of Convolutional neural network. A
typically CNN has divided into two different parts.
 Convolutional base, which has composed by stack of convolutional and other type of
pooling layers. In this way, it can easily identify primary goal is to generate specific
feature from blood cell image classification.
 Classifier, it has composed with multilayers which are established a strong connection
between them. The primary goals of classifier is to divide into small image blood cell and
easily detect the features. On the other hand, A fully connected with different layers
where neuron have establish connection with proper activation.
One important aspects of deep learning model is that when they can automatically learn entire
hierarchical features. So as handle to be computed features by multilayers. Generally, it can be
reused in the different domain while computed by last layers. The entire process may depend on
the data sets and various task. As a result, convolutional base of network especially from lower
to higher layers (Kim, Moon and Kang, 2019). During task execution, it is generally referred the
specific features whereas classifier part. Furthermore, it has pre-trained model for fulfilling
specific need and requirement. It will be starting to remove its original classifier and then add
new classifier that fit for specific purpose.
For image classification, it is standard approach to use the multiple stack of fully
connected layers which mainly followed by Softmax activated layers. It is helping in the output
layer where distribution of probability related result generation. It can be shown the possible
class label and then just require to classify image as per probable classes. Another type of
9

approach such as Global average pooling, it is based on the approach which are established the
connection with different layers on convolutional base. Sometimes, it is adding the pooling layer
and feed its outcome but it directly associated with the softmax activation layer. Linear support
vector machine is a type of approach which can improve the classification accuracy with the help
of training linear. It is mainly classified on the feature that will be extracted through
convolutional neural network.
Aim:
To investigate the Convolutional Neural Network and transfer learning to classify the
images by using python programming language.
Objective:
 To determine the concept of convolutional neural network and transfer learning.
 To identify advantage of convolutional neural network and transfer learning in the
extraction of blood cell image classification.
 To investigate the classification of blood cell image by using python programming
language.
Research Questions:
 What is concept of convolutional neural network and transfer learning?
 What are different advantages of convolutional neural network and transfer learning in
the extraction of blood cell image classification?
 What are classifications of blood cell image by using python programming language?
LITERATURE REVIEW
Theme:1 Determine the concept of convolutional Neural network and transfer learning.
Qi and et.al. (2019) A convolutional neural network is based on the learning technique
which mainly used to take input image and assign significant to different aspects in blood cell
images. In this way, it able to differentiate one from another. Sometimes, it is also known as pre-
processing which required for handling quality of image. While another method is to filters with
training convolutional has an ability to identify characteristics. CNN method can be categorised
into deep learning while identifying the difficulties in neural network which may overcome
through pre-training. Afterwards, it can be implementing layer wise pre-training in order achieve
the specific results or outcome. Convolutional neural network is a kind of network which help
for designing an effective structure effectively reduce the complexity in image recognition. This
10

type of technique will may include specific features such as adaptability, simple structure design
and pre-training parameters.
Generally, the structure of CNN may include two different layers, which is one of most
common features of extraction. Therefore, each neuron considers as input that connect with local
fields and extract from local features. The feature map is mainly used as sigmoid function as
activated functionality within convolution network and make features to shift invariances. CNN
is primarily used to determine the displacement and in another form of design two-dimensional
graphics. It is the most effective features that can help for identifying detection layer of
convolutional neural network by training information.
In this way, it plays important role in blood cell image classification where convolution
neural network technique uses to extract feature from raw blood cell image. The overall
architecture of CNN may include different layers such as fully connected layer, convolutional
layer and polling layer. In most of cases, convolutional layer computer with the neuron outcome
by estimating overall weight of input, adding bias to weight sum and then applying to filter
effectively. In this way, it is considered as one of important activation function with CNN
technique whether neuron should be activated. On the other hand, pooling layer will change to
reduce overall size image of blood cells. In this way, it will reduce the parameters number as
well as computations. The fully connected layers are containing a lot of neuron which directly
related to activation function from convolutional layers. In this process, it must use network
containing different layers which performed extraction while classified overall features. The size
of input is perfect to determine each layer in proper manner. The conventional layer is applying
filter option to categorise in different manner. In this way, it can be identified that it become
simplest technique to enhance level of contrast image and detecting overall distribution of blood
cell image pixels. In order to stretch the range of intensity value and calculate desirable outcome.
Byra and et.al (2019) argue that convolutional neural network is one of most common
approach in term of deep learning structure for image classification. CNN network become
popular after exceptional performance and identifying actual blood cell image quality. In
additional, it is similar to the other forward network in the input which are feeding through input
layer and processed. Moreover, Convolutional neural network has different layers which may be
including input which allows for capturing representative features from input data. Another step
11

of CNN concept is to utilise identical entire weights to process specific input and easily
synthesises its unique features.
Fahimi and et.al. (2019) transfer learning is based on the machine learning technique
where it can be developed model for reusing task in term of blood cell image classification. It
can be starts from initial step to last step. It is to be considered the most appropriate approach in
term of deep learning where it will be used pre-trainer model. It is mainly starting from hardware
vision where how they can recognise image through natural language processing. It is vast topic
which compute with resources to develop network model for identifying any problem through
image recognition. At that time, it can apply transfer learning technique to transfer entire weight
parameters that covered pre-trained on image recognition. It should be collecting large amount
data set which adopt to customise loss function. It allows for neural network to train and cover
faster speed in image recognition, classification. Hence, this process is mainly applied to detect
abnormal condition in cell which by means of segmentation as well as classification. As a result,
it has shown the 89% accuracy that’s why, method always proposed for further enhancement by
taking as a classifier. Yin and et.al (2020) said that transfer-based learning cell image
classification obtain the multiple patches which are exhausted in pixel wide sliding. However, it
also identified the major drawback whereas splitting the application on the basis of blood cell
classification. It will be generating large number of unwanted patches in different layers.
Although transfer learning and convolutional neural network have shows as good result
or outcome in terms of blood cell classification. Sometimes, it can be divided inti white blood
and red blood cells. It is very important to understand functionality of human immune cells and
then after it can be used method to divide range of blood cells into different manner.
Theme: 2 identify advantage of convolutional neural network and transfer learning in the
extraction of blood cell image classification.
Coulibaly, Kamsu-Foguem and Traore (2019) Convolutional neural network is now go
beyond every cell image and also relates to the specific problem in blood cells. In this way, it is
successfully recommending the CNN technique to process in blood cell image classification as
well as recognition. The advantage of convolutional neural network is that when it automatically
detects the features without supervision of human. It is the most powerful computational
technique which are becoming efficient to identify any type of problem within blood cells. It
benefits of special convolutional and also pooling multiple operations, it is performed different
12

parameter for purpose of sharing. It enables convolutional neural network to run hardware
devices which are making them attractive. This type of technique is dealing with most powerful
and efficient model which mainly performed as an automatic features extraction to easily achieve
the better accuracy in result generation.
Another advantage of convolutional neural network for stacking different functions
instead of use relu operations. It provides the better non-linearity and powerful in the model
design. There are large number of filters increases as go into overall network. It can be
implementing the spatial size of decreasing feature map while handling overall capabilities
within few minutes. The probability of deep learning which makes itself in different task with
datasets. Therefore, it is also increasing advantage of easily unlabeled information and data,
generating the accurate model in the form of different layers. It is beneficial technique which
help for improving overall performance and efficiency of result investigation while finding
specific problem and issues within blood cells.
Marzahl, Aubreville and Maier (2019) The Advantage of Convolutional neural network
always motivated by facts that help for capture relevant features from image at different level.
Sometimes, it also become similar to human brain. It is a type of weight sharing which easily
explore the multiple layer and represent in the form of image. In this way, it will be used the
convolution technique for calculating parameters. Clearly, it also analysed that become more
efficient in terms of complexity and memory management. Convolutional outperform on
recognize image processing task for classified into different. For instance, it is completely new
task which will consider good features in blood cell extraction.
Acevedo and et.al (2019) the advantage of transfer learning in the blood cell image
classification for reducing cost of computation. Initially, it is trainer on large image data set and
applied suitable task for result identification. Transfer learning is based on the machine model
which have nee trained for specific task but it easily changes the status of current situation or
conditions. The primary advantage is to trained, identify problem and used on different ways. It
drawn the knowledge which has been already applied in the new task management. It could be
simpler as training model to recognize accurate or correct image. In this way, it can be identified
that transfer learning is the most powerful tool which often to reliance large volume of data or
information. Unfortunately, advantage is to sort the data sets which increasingly prohibitive
expensive.
13

Jokhio and Jokhio (2019) another advantage of transfer learning in term of neural
machine where it will be used to explore cross-lingual model for resolving problem with lack of
resources. The several transfer learning approach to reuse model which pertained on high
resources. It also improved the overall performance during blood cell image recognition. It is
beneficial for analysis overall threats and problem with the help of transfer learning technique
because it easily found problem within blood cells. Another advantage of transfer learning
through deep learning and increasing the overall functionality for image recognition. As a result,
it has shown accurate result or outcome.
Training learning help for gathered data from multiple source which provide pre-trained
ConvNets along with fine-tuning policies which is better for identifying depth network analysis
from scratch. When it is targeting large amount data set in smaller than other data set. Transfer
learning is becoming efficient which enable to Inception network training without any problem
in regards of convergence and overfitting. Initially, it will be utilized the overall convolutional
and also fully connected with multiple layer from image models.
Theme:3 Investigate the classification of blood cell image by using python programming
language.
Sahlol, Kollmannsberger and Ewees (2020) Python programming language is based on
the general high-level powerful which mainly used for creating prototypes. It is one of most
common language that mainly used for image processing. It consists of multiple block that are
imported to python, considering the programming environment which is becoming flexible.
When after preparing blood sample on different slider of image that processed. Sometimes, it can
be done by using python interface and library. It should be considered the algorithm in the blood
cell image classification through multiple image processing techniques. It ensures that RBC
visible and possible to generate significant result or outcome. The image processing step is being
enhanced in term of quality management that being prepared for improving overall process.
Sometimes, it may have arisen problem in the blood cell image such as andillumination issues. It
should consider convolutional neural network technique to use feature of extraction that must
follow by descriptiondeals.
Jokhio and Jokhio (2019)Afterwards, it will be generating result in some Quantitative
information as per features for one class to another. Through python interface, it will count the
number of cells that covered by view of camera and completed by image processing algorithm,
14

but the actual unit of blood cells may be count which appears on specific documentation.
Therefore, it is obtaining the sample which included anticoagulant and diluted liquid,
automatically separate the cells and control the overlapping situation. The actual result of sample
was identified and verified with count large number of blood cell image and validate the
classification in proper manner. It may be processed original sample by using haematology as
analyser that will show a reading of 5.48 million cells. The image processing algorithm shows as
count of 5.20 million per cm. In this way, the proposed system will help for generating accurate
result and show error approximately 3.45% and accuracy of 98%. It is to be considerable for
taking into account with normal range of blood cell image and their effectiveness of proposed
system. The extraction of cell image obtained through python programming because it applied
image processing algorithm for counting blood cells. Sometimes, it also detecting the blood cell
whether after being tested. Therefore, it is suited the most that can easily utilized for increasing
efficiency of system. A comparative study with image from different sources which being tested
whether observed image through image processing algorithm. It is the best way to find out the
cells image effectively and efficiently. After dependencies, it is becoming installed and
information which are placed in accurate directories. In this way, it useful for running various
classification approach in command prompt. It is running the logistics regressing by
implementing extraction features, follow the command in top directory that containing python
code.
As per Albahar (2019) Python is mainly dealing with programming which help for
identifying classification in term of RBC, Platelets and WBC. By using detection algorithm, it
easily detects and find out difference between blood cells. It also accomplishes the accurate
result with RBC of 96% and WBC of 98% with minimum epochs. It can be applied the deep
filter through Convolutional neural network, require the suitable sample with average sizes and
generate multiple variances. In this way, it can derive to optimize convolution neural network
technique for giving accurate calculation in terms of segmentation output. On the other hand, it is
important concept for classified the blood cell image which are taken image processing
algorithm to compare with other cells. There are various types of multinomial logistic regression
which provide better performance with 98% test success and become easier for calculating the
blood cells image.
15

Rani and Kumar (2019) Python uses for identifying causes which is decrement and
increment of blood cells, utilized the certain protein through diagnosis process. However, it is an
essential to deep quantify in terms of analysis. Threshold method with Convolutional neural
network-based on layers classified, which are obtaining accuracy in different thresholds. Blood
cell classification accuracy level is become 90% and 89% for positive in image processing. It can
be identified the magnification image which has taken as source image. Here, it is combined with
Convolutional neural network through python with data augmentation. It is set of input image
which are given to train by using CNN. In additional, it also involves the data augmentation that
will create own data to reduce problem in quality of blood cell image. It is to be normalized
through batch and produced sample for applying in overall model. Furthermore, Dense layer will
be increased to get accurate result and achieve hyper parameter. In additional, It will be complied
with the image processing and programing. The set of data employed in training as well as
testing face by using transfer learning. In order to use trained data set through parameter values.
Sometimes, it automatically generates default values. However, it was done overall coding to
reduce size of image adversely affects attributes in blood cell extraction. The validity of blood
cell classification is relating to accuracy, sensitivity, specifying. Which only possible to use
python codes because it supports for categorized into different groups. In this way, it is helpful
for identifying better accuracy in result or outcome. At starting point, CNN model tried to select
particular cell image and classified through classifier softmax. Sometimes, it is generating
highest accuracy and obtained with 85% success rate. Another included extractor features which
is one of most common features with multiple classifiers.
RESEARCH METHODLOGY
It is based on the technique or procedures that used to identify, select, process and analyse
information about the topic or subject. The methodology allows the research to critically evaluate
study overall validity and reliability (Kumar, 2019). Research can be involved the activity that
finding out, in more or less systematic ways. On the other hand, research methodology is
considered as philosophical framework within performed the activities.
Research Method
It is a type of systematic plan for conducting research which has drawn on variety of both
quantitative and qualitative research methods. It is the specific tools, procedures that use to
collect or analyse information or data in related to research project.
16

Quantitative research method is mainly described and measure the level of occurrence on the
basis of calculations, numbers. It is entailing the collection of numerical data and also exhibiting
the specific view of relationship between different theories which easily predict the data on the
basis of numerical aspects (Kebritchi, McCaslin and Rominger, 2019). The quantitative method
examines the relationship between numerical instructions to measure variables with application
of statistical techniques.
On the other hand, Qualitative methodology is based on the pre-existing theories which
always support for gaining insights and also understanding the reason as well as motivations.
This type of method is able to express human emotions. The data must be often regarded as
providing the information on behalf of real-life situation. It is being able to understand and gain
more information through previous study.
The researcher has chosen Qualitative research method which help for gathering and
collecting large amount of data or information which are relevant to the research project. It helps
for research to understand experience, attitude, beliefs, interactions and behaviour of people. in
this way, investigator is integrated with the qualitative theories that supports in the intervention
study, gaining increased attention across the world. The qualitative method can be dismissed as
over simplifying the person experience which require to gain more understanding through
aggregate data. Investigator has studied about the different theories which are based on non-
quantifiable data to collect large amount of information effectively and efficiently.
Research Design
It is defined the overall strategy that applicable choose to integrate with multiple
components of research study. It is becoming coherent and logical manner. It must ensure that
will address or find out the research problem (Snyder, 2019). It also constitutes blueprint for
gathering or collecting data. It can be defined that research design as plan about what research
find suitable output or result. Research design can be categorised into different group:
exploratory and conclusive.
Exploratory research is that when it implies, intended to explore the research question
that mainly used in the project. This type of research is clarifying the exact nature of research
problem to be resolved. It has undertaken into considerations during analysis. The exploratory
research design is not providing the final conclusion but explore the research topic with the depth
level. On the other hand, conclusive design is another type of approach that help for identifying
17

accurate result. It mainly used the practical aspects to reach on the specific conclusion which
need to clearly define in proper manner. It provide the better way to quantify as well as verifying
the research finding.
The investigator has chosen exploratory research approach which is mainly used in
qualitative research method in order to identify the nature of problem and issues that occurs
during research study. It is helpful for researcher to improve their better understanding towards
the problem where they are applying exploratory to gain positive result or outcome. Furthermore,
researcher has adopted the technique to explore the research question, aim to provide final
finding in further expansion.
Research Philosophy
It is based on the knowledge, source and nature that simply belief about the way to
analyse, collect and use data in proper manner. Although, it has developed the certain idea,
thoughts, belief on the research project (Quinlan, Babin and Griffin, 2019). Researcher may
engage in the suitable knowledge through philosophy and consider as important part of case
study. Research philosophy mainly involves about formulation, aware about the belief,
assumption. It become easier process to clarify in research method. The research philosophy can
be classified into different types: Interpretivism, pragmatism, realism and positivism.
The investigator has chosen interpretivism philosophy which mainly used by interpret
elements in case study. Researcher is integrated the philosophy to the human-interest during
research analysis. In this way, it can easily assume to access reality because it is based on the
naturalistic approach in the information gathering or collecting. It is the most popular philosophy
in qualitative methods that’s why, researcher apply within case study to emerge with the
processes. Main reason of adopting interpretivism by researcher, when it helps for identifying
cross-cultural difference in the organization. In this way, researcher should be considered the
different factors that impact on leadership. Investigator has been studied with high level of depth
knowledge about the topic. It might be associated with high level of validity during data
collection. The study tends to be honest and trustworthy across the world.
Research Approach
It is based on the procedure and plan that mainly consists of different phases in regards of
assumption in detailed. In this way, it can easily analysis and interpret data in proper manner.
The research approach is being collecting information relevant to research project and also
18

addressed problem (Phillips and Ritala, 2019). In research study, approach is the important part
of scientific study which are dividing into formulation on the basis of hypotheses. Research
approach can be categorised into different ways: Abductive, deductive and inductive research
approach
A deductive approach is a concept that will develop a hypotheses idea through existing
theories and then planning for design in research project. In this way, it is linked with the
previous theories that applicable in case study. It might be tested the relation and obtain the
specific circumstances. On the other hand, inductive approach is based on the observation which
mainly proposed the result or outcome through research process. This type of approach is
including multiple pattern by using observation and also developed proper explanations. The
primary aim is to generate accurate, meaningful result from large number of data set. in this way,
it can easily identify pattern and relationship with other theory.
Abductive approach can be defined as set of process where address weakness within
research study. It also associated with the other approach effectively and efficiently. Sometimes,
it also considered as alternative option to overcome challenges, problem and issues within
research project.
The investigator has chosen inductive research approach in the case study because it
helps for researcher to identify the different pattern of information. Researcher use inductive
approach to start with analysis for starting point and also free in term of altering the path or
direction of case study. The researcher has adopted inductive approach to move any direction
where it can be generated specific result or output. It is also important to stress that inductive
approach doesn’t completely imply disregarding theories when it easily formulating the research
objective or questions. The approach is to generate the proper result while establishing a strong
relationship between different theories.
Sampling
It is a type of specific principle that applicable within research project to select members
those are participating in the case study. It is right way to identify the interest of people in
research study. There are large number of population’s shows interest in project but it is
necessary to select suitable members for participating in research processes. In order to identify
accurate result, output (Pheng and Hou, 2019). In another words, there are large number of target
population but investigator has no choice but it can study about the different elements within
19

population. it is representing with case study in order to reach the suitable conclusion. Sampling
method can be categorised into different type such as non-probability and probability.
The investigator has chosen the simple random technique within case study. This
technique is based on the straight forwards probability in term of strategy. it is to be consider for
selecting the particular sample in the large population for wide purpose. The researcher has
chosen simple random where they are likely to be select specific part because there are some
logics behind technique. It also removed the bias from the specific selecting process and also
representing the result of particular sample.
It is true when large sample size required within case study so that researcher able to
apply in proper manner. It also identified that simple random sampling easy for investigator to
understand theory but it is very difficult to perform in proper manner. The sample size is 40
employees.
Data collection method
It is a type of process of gathering and collecting information from authenticate source to
identify answer to the research issue or problem. In this way, it also tests hypothesis and also
evaluating the suitable result (Kumar, Mariano and Udani, 2019). in another way, data collection
process is an essential for gathering, measuring information or data on variables of interest. in
this way, it has established systematic fashion that evaluate one or more queries. There are two
type of collection technique such as primary and secondary data collection.
Primary data collection is a process for collecting large amount of data in proper manner.
The method is directly gathering data from authenticate sources which opposed to collect
information related case study. It involves face to face, interviews and observation.
Secondary data collection is another type of method in term of information choosing that
has already published in newspapers, journals, articles and portals. The data is available in the
different sources about the research field in studies. In another way, it also selecting the
secondary data to the study analysis while maintaining the level of project reliability as well as
validity.
The Scholar has chosen primary data collection method in the research study
perspectives. It helps for researcher to identify suitable information which always support for
generating result or outcome. Through qualitative method, researcher has used the questionnaires
method for collecting the data or information. It must ensure that high level of depth
20

understanding towards the relevant authenticate data. This type of method is useful for researcher
because it allows time to easily formulating research question and understanding the issues. It is
also important to have unused data and determine any information.
Data analysis
It is a type of method which mainly refers to the non-numeric data or information such as
text documents, video and images (Kumar, Mariano and Udani, 2019). In qualitative data
analysis, researcher has chosen thematic analysis for identifying suitable information related
research topic. The process is useful for researcher of evaluating data through analytical and
logical manner. Sometimes, investigator has examined the components of data gathered in
different steps. The important steps that are taken when research analysis is being conducted in
properly. In qualitative, researcher has been utilised the processes for purpose of analysis and
provide the better understanding towards interpretation.
Ethical consideration
It is an accumulation of principle and value that would address the specific research
question. Ethics are considered for reason and act as maintaining the rules and regulations. It is
also approving the conduct for believing or denying somethings. Ethical considerations can be
specified as one of most important part of research. There are different types of principles related
to ethical considerations.
 The research participants should not be subjected to harm in another ways.
 It also requires to respect for dignity of researcher and also give more prioritised.
 it should be obtained from other participants which give more prior to the case study.
 It must ensure that should be obtained proper information or data.
 Adequate level of confidentiality of research data or information should be ensured.
 It required to avoid the deception or exaggeration about research objective, aim must be
handled in proper manner.
 Any type of misleading data or information should be representing in primary data
finding which are avoided in proper manner.
Validity and reliability
Reliability is referred to the concern about the single observation by using authenticate
sources. There is no certain guard against the impact of other researcher. The reliability problem
or issue is that when it is closely associated with objectivity. If researcher adopt the subjective
21

approach in the research case study. In this way, it can be identified the reliability of work which
are going to be compromised in proper manner.
Validity is also defined the process for identifying requirement of scientific research
method that have been followed at the process of collecting research finding (Kumar, Mariano
and Udani, 2019). It is to be considered the validity which are compulsory requirement for
different case studies. Sometimes, it is an essential for researcher to measure validity of research
but there are not limited point followings:
 It should require the appropriate time scale that has to be selected.
 It can be applied the suitable methodology, taking into account the specific features,
characteristics in the research case study.
 The most appropriate respondent participate for sharing their own opinion, view point in
regards of experience. It is an essential for selecting the best choice among population.
 It must ensure that respondents are not be pressured in any ways, select the best choice
among large number of sets.
As per concluded that research validity and reliability can be eliminated whenever investigator
need to strive and also minimize significant threat during research analysis.
DATA INTERPRETATION
It is based on the process of making sense out of large data collection and further
processed (Kumar, Mariano and Udani, 2019). It may be represented in multiple formats such as
tabular, charts, graphs and other type of forms. Therefore, it needs to interpretation of different
kind of data or information during research project analysis.
Theme: 1 building a neural network where it gets input from layer as classified blood cell image.
Q1. Do you think that building a neural network where it gets input from layer as classified
blood cell image?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
10
20
5
5
Total 40
22

Table: 1
Figure 1
Data Interpretation:
In above table, it has summarized about the building a neural network where it gets input
from multilayers. It is mainly used for classified the blood cell image into different layers. In this
way, it become easier for identifying the image of blood cell and also categorized according to
their structure. As per discussion, it has found that maximum respondent agree on the statement
where they are agreed on the neural network, which help for supporting in the image
classification. Furthermore, it can be determined that 10 respondents are strongly agree, 20
respondents are agreed, 5 strongly disagree and 5 respondents disagree. In this way, it can easily
identify the opinion of each respondent in proper manner.
Theme: 2 convolutional neutral network become advanced technique that mainly used in the
blood cell image classification.
Q2. Do you think that convolutional neutral network become advanced technique that mainly
used in the blood cell image classification?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
12
10
10
8
Total 40
23

Table: 2
Figure 2
Data Interpretation
From above chart, it has been found that convolutional neutral network consider as
advanced technology in term of blood cell classification. It provide the better view about the
overall blood cells into multiple layers. As per analysis, many respondents are sharing their own
opinion on the basis of personal experience. In medical term, convolutional neural network
technique support for increasing efficiency and performance of blood classification. In order to
generate accurate result or output. Furthermore, it can be determined that 12 respondents are
strongly agree, 10 respondents are agreed, 10 strongly disagree and 8 respondents disagree. In
this way, it can easily identify the information about the convolutional neutral network.
Theme: 3 Estimation of red and white blood by using Convolutional neural network technique.
Q1. Do you think estimation of red and white blood by using Convolutional neural network
technique?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
15
10
10
5
Total 40
24

Table: 3
Figure 3
Data Interpretation
In above discussion, it has summarized that maximum respondent agree on the statement
where estimation of red and white blood by using Convolutional neural network technique or
method. Generally, normal blood cell contain both red and white blood cells so as require to
classify through CNN. In this way, it become easier for medical professional to compare blood
sample with one, predict the accordingly. As per analysis, many respondents are sharing their
own opinion on the basis of personal experience. Furthermore, it can be determined that 15
respondents are strongly agree, 10 respondents are agreed, 10 strongly disagree and 5
respondents disagree. In this way, it can easily identify the information about the convolutional
neutral network.
Theme: 4 Confusion matrix applicable in CNN technique for identifying color texture features of
blood cells.
Q4. Do you think that Confusion matrix applicable in CNN technique for identifying color
texture features of blood cells?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
12
14
10
4
Total 40
25

Table: 4
Figure 4
Data Interpretation
In above table, it has found that confusion matrix applied during convolutional neutral
network method for dividing blood cell into multilayers. It help for identifying the overall
picture, structure and texture of blood cells. The matrix is providing the blood cell design where
how they can easily divided on the basis of texture, colors. As per analysis, it can be determined
that 12 respondents are strongly agree, 14 respondents are agreed, 10 strongly disagree and 4
respondents disagree. In this way, it can easily identify the information about the convolutional
neutral network. Each and every respondent shows their opinion about the confusion matrix that
can easily identify the texture of blood cells.
Theme: 5 Is it beneficial to use convolutional neural network and transfer learning in the blood
cell image classification.
Q4. Do you think that beneficial to use convolutional neural network and transfer learning in
the blood cell image classification?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
12
11
10
8
Total 40
26

Table: 5
Figure 5
Data Interpretation
In above discussion, it has summarized that maximum respondent agree on the statement
where It is beneficial to use convolutional neural network and transfer learning in the blood cell
image classification. In this way, it can easily collect the information about blood cells and their
specific divisions. Convolutional neural network and transfer learning that help for predicting
about the extract of blood cells. As per analysis, many respondents are sharing their own opinion
on the basis of personal experience. Furthermore, it can be determined that 12 respondents are
strongly agree, 11 respondents are agreed, 10 strongly disagree and 8 respondents disagree. In
this way, it can easily identify the information about the convolutional neutral network.
Theme: 6 transfer learning tool help for reducing to cost of computational in extraction of blood
cell classification.
Q6. Do you think that transfer learning tool help for reducing to cost of computational in
extraction of blood cell classification?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
10
12
10
8
27

Total 40
Table: 6
Figure 6
Data Interpretation:
In above discussion, it concluded that transfer learning is best tool that help for
identifying the accurate result or outcome. As per analysis, maximum agree on the concept of
transfer learning that can be used in the medical term to extract large amount of statistical data
set of blood cells image and also reduce cost of computational. In this way, it easily classified in
different layers. Furthermore, it can be analysed that can be determined that 10 respondents are
strongly agree, 12 respondents are agreed, 10 strongly disagree and 8 respondents disagree. In
this way, each and every respondent share their opinion as per experience and show positive
review towards the use of transfer learning platform.
Theme: 7 neural network support for improving performance of blood cell image recognition
Q6. Do you think that Neural network support for improving performance of blood cell image
recognition?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
15
8
10
7
Total 40
28

Table: 7
Figure 7
Data Interpretation
From above discussion, it has summarized that maximum respondent agree on the
statement where Neural network support for improving performance of blood cell image
recognition. It is useful for gathering and collecting the large number of data in the form of
multilayers. Convolutional neural network support in term of image recognition so as implement
within the extraction of blood cell classification. Therefore, respondent share their own view
point about the neural network in term of blood cell image recognition. As per analysis, it can be
determined that 15 respondents are strongly agree, 8 respondents are agreed, 10 strongly
disagree and 7 respondents disagree. In this way, it can easily identify the information about the
convolutional neutral network.
Theme: 8 Extraction features representing as important step to classify the area, compactness,
multiple layer and distribution
Q8. Do you think that Extraction features representing as important step to classify the area,
compactness, multiple layer and distribution?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
15
8
10
7
29

Total 40
Table: 8
Figure 8
Data Interpretation
From above table, it has summarized about the Extraction features representing as
important step to classify the area, compactness, multiple layer and distribution. There are
maximum respondent agree on the features that applicable within convolutional neural network
for easily categorised into different manner. Furthermore, it has concluded that 15 respondents
are strongly agree, 8 respondents are agreed, 10 strongly disagree and 7 respondents disagree.
Each and every respondent share their own view point blood cell extraction which help for
managing large amount blood cell image. In order to identify the multiple layers and distribution
on the basis of structure or layout.
Theme: 9 Python use for identifying causes which is decrement and increment of blood cells,
utilized the certain protein through diagnosis process.
Q9. Do you think that Python use for identifying causes which is decrement and increment of
blood cells, utilized the certain protein through diagnosis process?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
10
10
5
5
30

Total 40
Table: 9
Figure 9
Data Interpretation
In above table, it has found that Python is the most efficient language which help for
different extraction of blood cell image. Sometimes, it also classified into decrement and
increment of blood cells, utilized the certain protein through diagnosis process. As per
discussion, maximum respondents are agree with the concept because python is based on the
high programming language. It can easily identify any type of defect that occurs within blood
cell image. That’s why, Convolutional neural network implement by using python language and
identify the disorder within blood cells. In order to identify that how much protein exists within
blood cell during diagnosis process. Furthermore, it can be identified that 10 respondents are
strongly agree, 10 respondents are agreed, 5 strongly disagree and 5 respondents disagree. In
this way, each and every respondent shows their own interest towards the python language and
also used for identifying accurate result or outcome.
Theme: 10 convolutional neural network technique reduce size of image adversely, affects
attributes in blood cell extraction.
Q 10. Do you think that convolutional neural network technique reduce size of image adversely,
affects attributes in blood cell extraction?
Option Respondent
 Strongly Agree 9
31

 Agree
 Strongly disagree
 Disagree
14
8
9
Total 40
Table: 9
Figure 10
Data Interpretation
In above table, it has summarized about the technique of convolutional neural network
which primarily used for reducing size of image adversely, it directly affects on the attributes of
blood cell extraction. As discussion, it concluded that maximum respondent give positive
response where they also believed. It is the most efficient technique or method that adversely
used within extraction of blood cell image classification. In order to divide multiple layers.
Furthermore, it can be identified that 9 respondents are strongly agree, 14 respondents are
agreed, 8 strongly disagree and 9 respondents disagree. In this way, each and every respondent
give their response through previous experience. In this way, it can easily collect or gather
information.
Theme: 11 Is possible when transfer learning technique to transfer entire weight parameters to
cover pre-trained on the image recognition.
Q 11. Is possible when transfer learning technique to transfer entire weight parameters to cover
pre-trained on the image recognition?
Option Respondent
 Strongly Agree
 Agree
9
14
32

 Strongly disagree
 Disagree
8
9
Total 40
Table: 11
Figure 11
Data Interpretation:
In above table, it has summarized about the transfer learning technique to transfer entire
weight parameters to cover pre-trained on the image recognition. There are maximum
respondents show opinion or response where they also believed. Transfer learning is most
efficient technique that considered the different parameters which mainly applicable within the
extraction of blood cell image classification. As per analysis, it can be identified that 9
respondents are strongly agree, 14 respondents are agreed, 8 strongly disagree and 9
respondents disagree. In this way, each and every respondent give their response through
previous experience. In this way, it can easily collect or gather information.
Theme: 12 by using detection algorithm, it easily detects and find out difference between blood
cells.
Q 12. Do you think that detection algorithm, it easily detects and find out difference between
blood cells?
Option Respondent
 Strongly Agree
 Agree
10
12
33

 Strongly disagree
 Disagree
8
10
Total 40
Table: 12
Figure 12
Data Interpretation
In above table, it has found that detection algorithm applicable within extraction of blood
cell image classification in the form of multiple layers. It is useful for detecting and finding the
difference between blood cells. Maximum respondents are providing their own experience about
the detection algorithm which are considerations as image recognition. In above discussion, it
can analyse that 10 respondents are strongly agree, 12 respondents are agreed, 8 strongly
disagree and 10 respondents disagree. Equally respondent shows both positive as well as
negative reviews. In this way, it can easily identified opinion of respondent in proper manner.
Theme: 13 Do you think that cross-lingual model explore to resolving problem related lack of
resources and increase performance of image classification.
Q 13. Do you think that cross-lingual model explore to resolving problem related lack of
resources and increase performance of image classification?
Option Respondent
 Strongly Agree
 Agree
10
12
34

 Strongly disagree
 Disagree
8
10
Total 40
Table: 12
Figure 13
Data Interpretation:
From above table, it has summarized about the cross-lingual model explore to resolving
problem related lack of resources and increase performance of image classification. This model
is basically included during image recognition which effectively explore the image of blood cells
into smaller units. In order to generate accurate result, it also provided the better facilities for
increasing overall performance in proper manner. As per discussion, it has found that each and
every respondent share their own experience through questionnaires in order to find out their
opinion regarding the cross-lingual model. Furthermore, it has concluded that 10 respondents are
strongly agree, 12 respondents are agreed, 8 strongly disagree and 10 respondents disagree.
Theme: 14 Do you think that Preprocessing the image with color mask to only select the nuclei
of blood cell image.
Q 14. Do you think that Preprocessing the image with color mask to only select the nuclei of
blood cell image?
Option Respondent
 Strongly Agree
 Agree
 Strongly disagree
12
12
10
35

 Disagree 6
Total 40
Table: 14
Figure 14
Data Interpretation
From above discussion, it concluded that preprocessing the image with color mask to
only select the nuclei of blood cell image. It is the better way to understand the concept for
analyzing and also categorizing the extraction of blood cell image. Each layers are representing
into different color so that it become easier for identifying the pattern of multilayers in proper
manner. As per discussion, it has analysed that each and every respondents share their own
experience through questionnaires in order to find out their opinion regarding the cross-lingual
model. Furthermore, it has concluded that 12 respondents are strongly agree, 12 respondents are
agreed, 10 strongly disagree and 6 respondents disagree.
Theme: 14 Blood cell classification shows the accuracy between 90% and 89% in image
processing.
Q 14. Do you think that Blood cell classification shows the accuracy between 90% and 89% in
the image processing?
Option Respondent
 Strongly Agree
 Agree
10
14
36

 Strongly disagree
 Disagree
10
6
Total 40
Table: 15
Figure 15
Data Interpretation:
From above chart, it has been found that Blood cell classification shows the accuracy
between 90% and 89% in the image processing. It provide the better view about the overall
blood cells into multiple layers. As per analysis, many respondents are sharing their own opinion
on the basis of personal experience. In medical term, convolutional neural network technique
support for increasing efficiency and performance of blood classification. In order to generate
accurate result or output. Furthermore, it can be determined that 10 respondents are strongly
agree, 14 respondents are agreed, 10 strongly disagree and 6 respondents disagree. In this way,
it can easily identify the information about the convolutional neutral network.
RESULT AND FINDING
It can be estimating the use of Convolutional neural network that can easily classified blood
cell image. It has been found that Blood cell classification shows the accuracy between 90% and
89% in the image processing. In this way, it can identified the better view about the overall blood
cells into multiple layers. As per analysis, many respondents are sharing their own opinion on the
37

basis of personal experience. Convolutional neural network technique support for increasing
efficiency and performance of blood classification. Furthermore, research project may be
processed original sample by using haematology as analyser that has been representing original
reading approximately 5.48 million cells.
The image processing algorithm shows as count of 5.20 million per cm. In this way, the
proposed system has been generating accurate result and show error approximately 3.45% and
accuracy of 98%. It is to be considerable for taking into account with normal range of blood cell
image and their effectiveness of proposed system. In the research project, it can be used the
Python as programming language that help for performing different task effectively (Mozaffar
and Lee, 2019). Therefore, CNN use Python for categorized the extraction of blood cell image
into multiple layers. It help for classified into decrement and increment of blood cells, utilized
the certain protein through diagnosis process. Researcher may agree with the concept because
python is based on the high programming language. It can easily identify any type of defect that
occurs within blood cell image. That’s why, Convolutional neural network implement by using
python language and identify the disorder within blood cells. In this study, it can be examined
the different data set which accessible through CNN.
The data set consists of total 12,422 images where each image contain 24 bit and resolution
320x230 pixels, extension of image in JPEG format. Afterwards, it also implemented the features
of extraction process and develop as a perfect architecture model. During extraction and
classification that automatically optimized effective manner. CNN is one of most efficient
method that help for update learning coefficient in multiple iteration. Sometimes, it can easily
adopt the learning parameter which measured rates on the basis of average first moments. In
another way, it can be designed with benefits, advantage of method or technique.
Classification result of Convolutional neural network architecture compiled by using Python.
CNN architecture Epoch Accuracy Lost rate
LeNet (Adam) 70.00 100 18.200
LeNet (sgd) 41.00 100 13.300
VGG-16 27.35 100 12.324
Alexey 84.47 100 12.099
38

Table: 1
The AlexNet architecture was compiled with the image processing software, python
programming language. It support for easily dividing into different large data sets. It has
employed in the testing phase and training model, which performed the task without any deep
transfer learning. It is not trained data set which are primarily used. In this study, the parameters
value of architecture applicable as a default value. However, it has considered the LeNet
architecture data entry size which was taken as 225x225 pixels. In this way, this process can be
done while reducing the size of image which adversely effects the attribute of extraction.
However, it become costly in context of time consumption and also used memory. The size is not
increased (Mozaffar and Lee, 2019). The large amount of data set is complied with other type of
GPU support. Initially, it has been set up the software and added more features of hardware used.
It is basically used the graphics card, 8GB RAM, and other 2.5 preprocessors. In research
project, it is related to the validity, sensitivity, accuracy which are mainly specifying the overall
image pattern in different manner. Initially, the primary stage is that when it has designed CNN
model and selected. Generally, it is leading the overall architecture of convolutional neural
network such as LeNet, AlexNet and VGG-16. These are different architecture that represented
the classifiers. As per estimation, it can be calculated the highest accuracy which obtained with
85% of success rate through CNN (Umapathy and et.al., 2019). Therefore, AlexNet architecture
is becoming selected as primary feature of extraction and classified into different manner. In
another phase, it mainly used the AlexNet architecture model with the help of Matlab software. It
provide the softmax classifier whereas 30% of large data set and tested the data. In this way, it
has identified that 70% as training data. For another purpose, classifier that mainly classified 10
fold as cross validation which applicable in data set.
On the other hand, AlexNet architecture must be compiled by using Matlab interface while
extracting the blood cell image classification. As per discussion, it has found different classifiers
such as KNN, DT, LDA, SVM and QDA. The success rate with different classifiers
approximately 98%. As a result, it has shown the confusion matrix which always obtained the
value from different classifiers. Apart from that success rate was estimated or calculated through
39

confusion matrix metric values into different classes. Sometimes, it also obtained the curve and
graphs. In another step, Attributes represented in this phase where QDA classifier are used. It is
to be consider as classifications process which can performed by technique. As per estimation,
there will be chances of 84.7% success which easily obtained and achieved. In order to examine
the multiple number of attributes which automatically decreased (Mozaffar and Lee, 2019).
Afterwards, it has found that classification process reduced from 97% to 85%. On the basis of
CNN which mainly increased success rate of blood cell image on the existing cells through
application. Pixels on the data set by using iterative clustering algorithm. In research study, it can
use the method within convolutional neural network but they are contributed positive way to
identify the accurate result or outcome (Umapathy and et.al., 2019). It has seen that classification
of large set become time consuming when it has increased the data set. In this way, it can easily
compared with other studies through existing data or information.
CONCLUSION
As per discussion, it has concluded the result or finding of blood cell extraction which
become easier to divide into different multilayer cells. It has been observed that acquisition of
image that could be arranged in short terms and also wide spread stress fibres as long bundles. It
either arranged in different actin filaments that were distributed evenly through cell body.
Convolutional neural network technique, which composed by pooling layers that support in
blood cell classification. The primary goal is to generate the specific features from image
processing. It has been taken as explanation of convolutional and multilayer structure or layout.
As per analysis, it can be used the different classified which usually composed by connected
layers so that it become easier for identification of blood cell image. In order to detect the
connected layer because each node establish connection with centralised neuron that has fully
connection. Afterwards, it automatically activated different layers in quickly manner. As per
discussion, it has found that deep learning important factor that support overall structure models.
In this way, it easily learn about the hierarchical feature. It means that computed by multi layers.
Generally, it can be reused in the different problem domain. Sometimes, it depend on the
situation where it consists of large number of data sets but it easily classified with the help of
convolutional neural network. Data analysis has summarised about the different concept related
convolutional neural network because it help for identifying structure, colours, texture of blood
cell image. In term of medical, it is the most efficient method extraction of blood cell image
40

classifications. Furthermore, it can be identified the structure of CNN may include two different
layers, which is one of most common features of extraction. Therefore, each neuron considers as
input that connect with local fields and extract from local features. Convolutional neural network
is a kind of network which help for designing an effective structure effectively reduce the
complexity in image recognition. This type of technique may include specific features such as
adaptability, simple structure design and pre-training parameters. In this way, it has proposed
blood cell classification method such as CNN which are majorly contribution towards the cell
divisions. It is best way to knowledge about application that combined with different model. The
combination model can be effectively implement spatial characteristics, features and temporal of
information or data. In order to achieve the better result or outcome. On the other hand, it is also
used the pre-trained method in CNN model which has chosen weight, closely interrelated to local
optimum. However, it keep record the high gradient wide range that are effectively collect large
amount of data or information. As per analysis, it can be determined the neural network which
capture dynamic information in the serialized format by hiding the connection between nodes.
Sometimes, it can be classified data and also forwards neural network, even also store the data in
the format of arbitrarily. Furthermore, it also introduced CNN with proper function to generate
specific output or result. In additional, it also extracted the features, characteristics from CNN.
The proposed technique adjusted the loss of function and activation from CNN, it must
require to optimise the overall multilayer into network structure. Each and every nodes are fully
connected with each other. That’s why, it help for dividing into different cells. So as helpful for
identifying the overall structure, layout of blood cell image. After implementing (convolutional
neural network) CNN that assume as the spatial size of minimizing feature map while properly
acquiring the data through high capabilities within few minutes. It has been calculated the
probability of deep learning which makes itself in different task with datasets and also converted
into specific layers which define division of cells nodes. On the other hand, it has concluded that
confusion matrix applied at the time convolutional neutral network technique, which can easily
dividing blood cell into multilayers. It help for identifying the overall picture, structure and
texture of blood cells. This type of matrix that has provided accurate extraction of blood cell
design, classification of different layers where how they can easily divided on the basis of
texture, colors.
41

As per evaluating the performance of convolutional neural network which mainly including
AlexNet, Xception, DenseNet and other ResNet-50. These are helping for extract the feature
from uninfected blood cells. The models has performed the task to optimize the hyper-parameter
by randomized grid search. Initially, it has identified wide search range and successful evaluate
learning rate. However, instantiated the CNN part in term of pre-trained as well as trained that
are established the connection between them. Sometimes, it automatically determine the different
layer during extraction and improve the entire performance in proper manner. In some situation,
it is evaluating the performance of pre-trained convolutional neural network in context of
accuracy. It is helping for identifying overall architecture model and weight of pre-trained CNN.
In this study, it has applied the level of algorithm which can easily detect or segments the blood
cell image.
The initial step is to detect cell where they can applied different scale based filter to identify
centroids of blood cell neuron. In order to generate as cell segments within multiple level of
active contour framework. Therefore, it continue to evolve the multiple cell boundary and
opening the operation during post processing. It has to remove any type of false detection object
such as artifacts that uses particular size of cell. Furthermore, it also filtered out through one to
one concept as consider cell ground annotation since extraction of blood cell classification
process have been evaluated after cell detection through manual point wise annotation. To do
applied the one to one matching process because it easily segmented different cells and also
checking the manual ground trust into regions. If in case, it has identified the point in region
which automatically counted the positive result or outcome. Another solution is to measure the
applied performance metric which customised with pre-trained models. Afterwards, it can
identify the optimum value respectively. The success of learning rate become optimum which
customised the convolutional neural network trained effectively and efficiently. The fully
connected layer identified from the exception which could be selected its extraction features
during blood cell image classification.
The measurement parameter metric has shown the performance that could be achieved by
pre-trained model in response to classify uninfected cells. Therefore, it is evaluating the
performance by extracting feature from multiple layers in the process while optimising the layers
of extraction features. It should be underlying the data or information. The different layers that
gave best result value from performance metrics and shows the result obtained by extraction.
42

While performed data analysis, it has been observed that result or outcome significantly
categorised, signify the data. In this way, it has opted to use the concept of non-parametric which
are automatically consolidate results. Researcher has observed that overall performance between
convolutional neural networks, revealed the significance between pre-trained and trained
customised model.
43

RECOMMEDNDATION
After successful completion of discussion related the Convolutional neural network and
transfer learning models which always support for categorising blood cell image into different
layers. It has needed to implement extraction features that help for initialising starting phase
where it applied concept or framework in proper manner. Furthermore, it has identified the
multiple layers which mainly divided on the basis of colors, texture, structure and layout. In this
way, it become easier for identifying any type of uninfected diseases within blood cells
(Umapathy and et.al., 2019). That’s why, it mainly focused on the CNN technique for
implementing during image recognition process and dividing into different layers in significant
format. In the study, it has recommended that segmentation of leukocyte by using Snake
algorithm. It is primarily used to feature of extraction into vector form 0 to 1 with normalization.
So that it become easier to reduce duplication whereas it generated an accurate result or outcome.
During evaluation process, it has made up the 1200 imagery group that consists of different
classes and generate 98% of correct segmentation result. Sometime, it has not enough so that
recommend to classify image while implementing the deep learning models because it consists
of Imagenet architecture. It has proposed the method for identification of different blood cells.
Afterwards, it has analysed the behaviour, nature parametric datasets in the range of multi-
dimensional. It can be possible when applied the support vector machine. In most of cases, it has
suggested find out result and also used to consist into different stage of blood cell image
classification (Mozaffar and Lee, 2019). During first stage, it has recommend to use pre-
classification with the help of back propagation algorithm. On the other hand, it has been
presented as the hybrid model which supported by vector machine. In order to reduce the level of
detection problem in extraction of blood cell image classification. Thus, primary aim to reduce
the negative perspectives. It has proposed that automatic thresholding algorithm for segmenting
the blood cell image and also enhanced the image recognise processing.
The recommendation is that when classifier used to divide into blood cells and also obtained
the higher accuracy result. Furthermore, it has suggested that image processing technique which
help for classified colors transformation. Edge detection, cell classification and image
fragmentation. It is an essential for process for maintain and control overall extraction of blood
cell image. In most of cases, the diseases has found through blood cell because they can
identified mismatched condition during blood cells so that it easily detect the problem. In the
44

classification of blood cells, geometric properties of image that are mainly used as input
parameters of artificial intelligence algorithm. It is also another important method which always
suggested to improve overall image accuracy while it might be representing in the different
format such as edge extraction and histogram equalization (Umapathy and et.al., 2019). The
geometric properties are classified into specific image and happens classification process. There
are approximately 100 type of different blood cell image which primarily used for purpose of
experiment. In order to understand the classified image and categorised on the basis of density,
texture properties, shape and other structure of blood image. Apart from that it has proposed as
simple process on the basis of colors information, features (Umapathy and et.al., 2019). As initial
step, when two different stage classified process which already contain the leukocyte cell
boundaries and nuclei. In second step, they are implementing the properties which obtained from
the linear discriminant analysis model. It also used the dimensional properties, function through
morphological features. In this way, it easily automatic classification of blood cells. So as require
to test extraction of blood cell which implemented the algorithm as per selected properties and
also obtained the accurate information related rate of classification.
It has been tested the algorithm which are obtaining an accurate result or outcome.
Sometimes, it also selected the feature such as nucleus area, ratio of nucleus into different cell
areas, nucleus area and also identifying average colors of cytoplasma. The ratio of each nucleus
areas to its environment, nucleus circularity and other members. During image segmentation, it is
important for applying better approach which can identify the suitable Convolutional neural
network. In the classification testing, it is primary used to generate the highest accurate
identification rate as 99.12%. In this way, it has focused on the classification into different
segments which obtained from blood cell and also compared with other algorithm. It has applied
the geometrical feature that obtained from different image, used as input parameter to generate
specific output. In this way, it has organised as follows: morphological feature in find out the
category of blood cells. The image based extraction happens with the help of machine learning
technique. As mention to handle the control and maintain the blood cell image classification into
smaller units.
Apart from that it has recommended that perform the blood tests through CNN while
analyzing the blood cell rely on manual counting patterns. It display the low efficiency and
involve string subjectivity. Therefore, automated method maintain cytometry flow which become
45

costly and cannot be classified into proper manner. In recently, it has been advised that use
image processing as well as pattern recognition technique because it is cheaper than other
method (Umapathy and et.al., 2019). In additionally, these type of processes can be divided into
wide range of application. Initially, it has increased problem but afterwards, it has segmented the
accurate image through nucleus, cytoplasm. These are uniform colors and shows as illumination.
Similarly, it has divided the colors pattern in both nucleus and cytoplasm.
REFLECTCTION
From the above study, I can say that it increased my knowledge and skills regarding
Convolutional (Zhao and et.al.,2019) Neural Network method. I came to know that how this
technique plays an important role. With the help of this technique doctors and professionals can
make their classification of blood cell easy. It can save their time of completing this process
which can improve their productivity and healthcare services in blood cells regards. I also knew
about the importance of literature review which researcher developed. With the help of literature
review, we can get experts different views on the main topic. These difference views directly
help us in accomplishing our main goal. In this study, the main aim of developing research was
to analyse the Convolutional Neural Network and transfer learning for classifying the images by
using python programming language. With the help of experts views on some questions I get to
know some advantages of Convolutional neural network technique and how it can be useful in
blood cell image extraction. I also came to know by literature review that how python
programming language can help doctors and professionals in classification of blood cell image.
So, I can say that literature review chapter helped me a lot in increasing skills and knowledge
regarding this technique and knowing the importance of blood cell classification. Further, I can
say that by developing this research I came to know that this technique has several features
which include: ability of adaption or flexibility, pre-training parameters and simple structure
design. Structure and design of this system is one of the main features which makes it able to
extract and classified images of blood cells. Its other layers also make it able to decrease overall
size image of blood cells. One of the main thing which presents in its connected layers is
neurons. It is the only thing which directly related to activation function from Convolutional
layers.
In addition, I can also say that Convolutional neural network as well as transfer learning
have several advantages in blood cell image extraction as well as classification. Some of the
46

features and advantages which I came to know by developing this study include: one of the main
advantage of this system or technique is it does not require human supervision as it has ability to
detect all features at its own. By making an effective use of this CNN system, doctors have
become able to detect or identify any type of problem occur within blood cell. It is important for
doctors to identify the main nature and type of disease. Without knowing the disease and the type
of problem in blood cell, they cannot provide effective healthcare services and treatment which
can make health of patients poor. So, for providing the best treatment and improving health of
people, it is important to have knowledge of the type of problem within blood cell and this can
be identified by this system. So, I can say that Convolutional neural network technique is a vital
for the success and plays an important role. By conducting this research, I can strongly can say
that Convolutional neural network system does not have fictitious features as all its features and
advantages are being proved by strong and real facts. From one of its facts, I also came to know
that it works as like human brain which makes it able to analyze all things of blood cell image at
different level (Shu and et.al., 2019). It is also a type of weight sharing which can be used for
different purpose such as: calculating parameters, memory management and recognizing image
processing.
Rather, Convolutional neural network technique, transfer learning in the blood cell image
classification also plays an important role. This transfer learning allows doctors and all
professional in healthcare department to reduce cost of computation by which they can increase
their productivity. This learning is mainly based on machine which is trained by people for
completing some specific task. And as per the training it performs in an exact manner and save
time of people. Bu I can also say that it can change the status of current condition. By research
on the topic and developing this research, i also came to know that one of the main purposes of
providing training to this transfer learning system is to make it able to identify or detect problem
and allow solutions for this problem in an efficient manner. Training also made it able to
recognize accurate blood cell image. This transfer learning model does not only identify the main
problem and the type of disease but it also helps people to resolve problem by exploring cross
lingual model (Han, 2017). So, this feature of detecting problems as well as providing
appropriate solutions can improve productivity and overall performance at the time of
recognition of blood cell image. I can say that transfer learning system can increase functionality
of image recognition and which can improve health of people. So, from all identified advantages
47

which I knew I can say that both transfer learning and that Convolutional neural network
technique plays an important role in classification of blood cell image.
In addition, I can interpret that this study increased skills and knowledge in the context of
that Convolutional neural network technique, transfer learning as well as python programming
language. In the context of python programming language I can say that it is mainly used for
creating prototypes. It is considered as one of the most important and effective language which is
mainly used for image processing. The main usage of this language and its multiple blocks is to
consider programming environment and analyze it. In this context, I also came to know that
people in hospitals who prepare blood samples on different sliders can make an effective use of
python programming language and interface to precede it effectively. It works as like an
algorithm in blood cell image classification by using different image processing techniques. It
also generates accurate results and also helps doctors in identifying errors in the process of blood
cell classification which improves overall productivity and performance and improve image as
well. So, I can say that this language plays an important role in classification of different types of
blood cells like white blood cells, red blood cells and platelets.
Further, I also learned the increasing importance of research methodology. In the whole
research, RM plays an important role because it consists of several elements. All these elements
allow researcher in identifying the best form in which they can gather information and data
relevant to the topic. One of the best type of research also allows scholar to complete and follow
further elements of research such as selecting the research design, approach and philosophy. By
selecting one of the best types of research, scholar can accomplish its goals by gathering relevant
and accurate information. In addition, in RM, I also came to know importance of data collection
and data analysis step. Data collection states sources and ways of collecting information and
without having proper and accurate information, it is not possible to even complete the research
and accomplish goals. So, selecting the best type between primary and secondary, researcher can
get all objectives, the main aim as well as improves its image in the eyes of viewers. Data
analysis is one of the main elements which play an important role in research. This element and
function allows researcher in making all collected information accurate and usable. So, I can
strongly say that RM is one of the main parts of research which needs to be performed at any
cost.
48

Lastly, in the last step of result and findings, I came to know that different people have
different views and by majority of answers and views I learned several new and actual things
about the main topic and blood cell image classification. From the data analysis chapter and
majority of views of people I came to know that neural network plays a vital role in classification
of blood cell image as they also said that it has now become one of the advanced technologies.
Majority of participants and sample also stated that transfer learning technique can definitely
help doctors in decreasing cost of computational in blood cell extraction. In addition, they also
said in the context of neural network that it can improve overall performance of blood cell
recognition which is one of the main features. I also came to know that Convolutional neural
network technique reduce size of image adversely which directly affects attributes in blood cell
classification of extraction because majority of people gave this answer and agreed on this. They
also believe that transfer learning, Convolutional network and python and used language playing
important role. So, from the above study, overall I can say that detection algorithm can easily
find out difference between blood cells and plays vital role (Foga and et.al., 2017). All these
systems discussed in the study are considered advanced technology and helping out to doctors in
saving their time and increasing performance as well.
According to me, it become great experience which acquire information related the
convolution neural network which mainly supporting to identification of blood cell images. I
have developed my own knowledge and gain information where CNN technique calculate
accurate result or outcome. Furthermore, I have believed that pre-trained and trained learning
technique generate the significant model which are generating the blood cell images. In order to
identify any type of defect within blood sample. Through research study, I have improved the
own knowledge which require to understand the concept of CNN, transfer learning in medical
terms because they are implementing as efficient method for controlling process in extraction of
blood cell image classification. I have identified the different method which useful for dividing
blood cell image on the basis of colors, structure, and texture. As I know that it is important for
purpose of blood cell classification. I have seen that many medical professional use this concept
for finding diseases through blood cells. In this research study, I have briefly understand the
overall concept of extraction of blood cell image classification by using convolutional neural
network technique. I have studied about many thing which acquire a lot of information or data.
Sometimes, I feel that CNN technique used as framework for increasing the overall performance
49

and efficiency of blood cell image classification. Sometimes, it may be provided that how to
improve overall image processing. As per discussion, I have identified significant role of python
programming language. It is to be consider as efficient process that controlled overall image
processing. In my personal experience, python is based on the high level computer language so
that it directly associated with the convolutional neural network during image recognition. In this
way, I have understand the step of image recognition where it support to generate more accurate
image pixels. Furthermore, I have improved own depth knowledge in term of blood cell image
recognition method while enhancing expansion in term of medical fields.
50

REFERENCES
Book and Journals
Acevedo, A. and et.al., 2019. Recognition of peripheral blood cell images using convolutional
neural networks. Computer methods and programs in biomedicine. 180. p.105020.
Albahar, M.A., 2019. Skin lesion classification using convolutional neural network with novel
regularizer. IEEE Access. 7. pp.38306-38313.
Byra, M. and et.al., 2019. Breast mass classification in sonography with transfer learning using a
deep convolutional neural network and color conversion. Medical physics. 46(2).
pp.746-755.
Coulibaly, S., Kamsu-Foguem, B. and Traore, D., 2019. Deep neural networks with transfer
learning in millet crop images. Computers in Industry. 108. pp.115-120.
Fahimi, F. and et.al., 2019. Inter-subject transfer learning with an end-to-end deep convolutional
neural network for EEG-based BCI. Journal of neural engineering. 16(2). p.026007.
Foga, S. and et.al., 2017. Cloud detection algorithm comparison and validation for operational
Landsat data products. Remote sensing of environment. 194. pp.379-390.
Han, X., 2017. Automatic liver lesion segmentation using a deep convolutional neural network
method. arXiv preprint arXiv:1704.07239.
Islam, J., Mubassira, M. and Das, A.K., 2019, February. A Speech Recognition System for
Bengali Language using Recurrent Neural Network. In 2019 IEEE 4th International
Conference on Computer and Communication Systems (ICCCS) (pp. 73-76). IEEE.
Jokhio, F.A. and Jokhio, A., 2019. Image Classification using AlexNet with SVM Classifier and
Transfer Learning. Journal of Information Communication Technologies and Robotic
Applications. pp.44-51.
Jokhio, F.A. and Jokhio, A., 2019. Image Classification using AlexNet with SVM Classifier and
Transfer Learning. Journal of Information Communication Technologies and Robotic
Applications. pp.44-51.
Kebritchi, M., McCaslin, M. and Rominger, R.A., 2019. Exploring a Research Methodology
Group Framework for Nurturing Research Methods and Designs: An Action Research.
Kim, J., Moon, J. and Kang, P., 2019. Recurrent inception convolution neural network for multi
short-term load forecasting. Energy and Buildings. 194. pp.328-341.
51

Kumar, A.H., Mariano, E.R. and Udani, A.D., 2019. Medical education research methodology:
accuracy and design. Regional Anesthesia & Pain Medicine.
Kumar, R., 2019. Research methodology: A step-by-step guide for beginners. Sage Publications
Limited.
Loganathan, V., 2019. EXTRACTION OF BLOOD CELL IMAGE CLASSIFICATION USING
CONVOLUTION NEURAL NETWORK.
Marzahl, C., Aubreville, M. and Maier, A., 2019. Classification of Leukemic B-Lymphoblast
Cells from Blood Smear Microscopic Images with an Attention-Based Deep Learning
Method and Advanced Augmentation Techniques. In ISBI 2019 C-NMC Challenge:
Classification in Cancer Cell Imaging (pp. 13-22). Springer, Singapore.
Mozaffari, M.H. and Lee, W.S., 2019. Bownet: Dilated convolution neural network for
ultrasound tongue contour extraction. arXiv preprint arXiv:1906.04232.
Pheng, L.S. and Hou, L.S., 2019. Research Design and Methodology. In Construction Quality
and the Economy (pp. 89-111). Springer, Singapore.
Phillips, M.A. and Ritala, P., 2019. A complex adaptive systems agenda for ecosystem research
methodology. Technological Forecasting and Social Change. 148. p.119739.
Qi, Q. and et.al., 2019. Personalized sketch-based image retrieval by convolutional neural
network and deep transfer learning. IEEE Access. 7. pp.16537-16549.
Qiu, W. and et.al., 2019. Multi-label Detection and Classification of Red Blood Cells in
Microscopic Images. arXiv preprint arXiv:1910.02672.
Quinlan, C., Babin, B. and Griffin, M., 2019. Business research methods. South Western
Cengage.
Rani, S. and Kumar, P., 2019. Deep learning based sentiment analysis using convolution neural
network. Arabian Journal for Science and Engineering. 44(4). pp.3305-3314.
Sahlol, A.T., Kollmannsberger, P. and Ewees, A.A., 2020. Efficient Classification of White
Blood Cell Leukemia with Improved Swarm Optimization of Deep Features. Scientific
Reports. 10(1). pp.1-11.
Shu, X. and et.al., 2019, April. White Blood Cell Classification Using Quantitative Phase
Microscopy Based Deep Learning. In Bio-Optics: Design and Application (pp. DT3B-
3). Optical Society of America.
52

Snyder, H., 2019. Literature review as a research methodology: An overview and
guidelines. Journal of Business Research. 104. pp.333-339.
Srivastava, S. and et.al., 2019. Optical Character Recognition on Bank Cheques Using 2D
Convolution Neural Network. In Applications of Artificial Intelligence Techniques in
Engineering (pp. 589-596). Springer, Singapore.
Umapathy, A. and et.al., 2019, January. Image Processing, Textural Feature Extraction and
Transfer Learning based detection of Diabetic Retinopathy. In Proceedings of the 2019
9th International Conference on Bioscience, Biochemistry and Bioinformatics (pp. 17-
21).
Yin, T. and et.al., 2020, January. Bridging the Gap between Laboratory and Field Experiments in
American Eel Detection Using Transfer Learning and Convolutional Neural Network.
In Proceedings of the 53rd Hawaii International Conference on System Sciences.
Zhao, D. and et.al., 2019. Relation path feature embedding based convolutional neural network
method for drug discovery. BMC medical informatics and decision making. 19(2). p.59.
53

Questionnaires
Q1. Do you think that building a neural network where it gets input from layer as classified blood
cell image?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q2. Do you think that convolutional neutral network become advanced technique that mainly used
in the blood cell image classification?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q3. Do you think estimation of red and white blood by using Convolutional neural network
technique?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q4. Do you think that Confusion matrix applicable in CNN technique for identifying color texture
features of blood cells?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q5. Do you think that beneficial to use convolutional neural network and transfer learning in the
blood cell image classification?
 Strongly Agree
 Agree
54

 Strongly disagree
 Disagree
Q6. Do you think that transfer learning tool help for reducing to cost of computational in extraction
of blood cell classification?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q7. Do you think that Neural network support for improving performance of blood cell image
recognition?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q8. Do you think that Extraction features representing as important step to classify the area,
compactness, multiple layer and distribution?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q9. Do you think that Python use for identifying causes which is decrement and increment of blood
cells, utilized the certain protein through diagnosis process?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q 10. Do you think that convolutional neural network technique reduce size of image adversely,
affects attributes in blood cell extraction?
 Strongly Agree
 Agree
 Strongly disagree
55

 Disagree
Q 11. Is possible when transfer learning technique to transfer entire weight parameters to cover pre-
trained on the image recognition?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q 12. Do you think that detection algorithm, it easily detects and find out difference between blood
cells?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q 13. Do you think that cross-lingual model explore to resolving problem related lack of resources
and increase performance of image classification?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q 14. Do you think that Preprocessing the image with color mask to only select the nuclei of blood
cell image?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
Q 15. Do you think that Blood cell classification shows the accuracy between 90% and 89% in the
image processing?
 Strongly Agree
 Agree
 Strongly disagree
 Disagree
56

57

58

59