Literature Review: Mixed Reality Visualization in Image Guided Surgery

Verified

Added on  2024/06/28

|80
|9595
|460
Literature Review
AI Summary
This document presents a literature review focusing on the application of mixed reality visualization techniques in image-guided surgery. It includes APA-formatted references and citations, journal rankings, and keywords for each reviewed paper. The review analyzes various solutions, techniques, algorithms, models, tools, and frameworks used in the field, detailing their goals, problems addressed, components, and processes. It critically evaluates the advantages and disadvantages of each approach, along with validation criteria, input/output data, and the research's value to the project. The document also features a review of a paper on artificial intelligence in healthcare, discussing its techniques like deep learning and machine learning, and their impact on medical decision-making. The aim is to provide a comprehensive overview of current research and identify potential areas for improvement and innovation in the domain of image-guided surgery using mixed reality visualization. Desklib provides this and many other solved assignments to help students.
Document Page
Literature Review (Secondary Research) Template
Student Name & CSU ID

Project Topic Title
<Technology, Technique, Domain; e.g. Mixed Reality Visualization in Image Guided Surgery>
NOTE: Please you need to use YOUR OWN WORDS in writing this template.

Your Literature Review Should be in Scope and MUST Address all Your Project's Questions
You should ONLY use CSU library, and Google search is NOT allowed. The papers you select should be in last 3 years. If you are in 2018, then
you need to collect 2018, 2017, and 2016.

We encourage you to search for Journal papers rather than conference papers as it will give you more details.
Check the Journal ranking (Q1, Q2, …etc) of the journal based on uploaded excel sheet in interact.
Sample with an Explanation of Each Section

Reference in APA format that will be in

'Reference List'

<
Author's surname and first name, year of publication, Paper topic, Journal name, volume no., issue no..,
pages range>.

e.g.
Kersten, M., Gerard, I., Drouin, S., Mok, K., Sirhan, D., Sinclair, D., & Collins, D. (2015). Augmented
reality in neurovascular surgery: feasibility and first uses in the operating room.
International Journal of
Computer Assisted Radiology and Surgery
, volumne 3, issue 2, p. 1823-1836.
Citation that will be in the content
<Author's surname and publication year>
e.g. (
Kersten et al., 2015)
URL of the
Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
<place the URL of the paper>
<Use the 'Journal Ranking' excel file, and
check the level of the journal, and place it

here>

<the keywords are available in the journal after abstract

section>

1
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
The Name of the Current Solution
(Technique/ Method/ Scheme/

Algorithm/ Model/ Tool/ Framework/ ...

etc )

The Goal (Objective) of this Solution &

What is the Problem that need to be solved

What are the components of it?

Technique/Algorithm name:

<Suppose your project topic is related to

visualizing techniques, then you need to

find the name of the visualization

technique that is proposed by (used by) the

author in each paper you review. e.g.

Energy Functional Technique for

visualization>

Tools:

e.g. MRI Scan, Stereo camera

Applied Area:

<Suppose your domain is visualizing the

anatomy of the body during the surgery,

you need to mention which type of surgery;

e.g. Liver Surgery>

Problem:

<What is the
problem in each paper you
review that
the author wanted to solve. You
can find it in the first paragraph in the

abstract>

Goal:

<
Why the author wanted to solve the problem.
what is his goal. What is the value of solving

him this problem. You can find it in the first

paragraph in the abstract
>
<What are the components of the system/model
. Place the
components as a bullet points only without any explanation.>

The Process (Mechanism) of this Work; The process steps of the Technique/system

<what are the process steps/stages of the proposed solution. Use
green colour for the step that shows the author contribution, and gives the best feature. What
is the step that give the
limitation (Use red colour for this step) >
The good features and limitations in each work, will based on YOUR ANALYSIS (Your Critical Thinking). You need to
Identify the problems of the
technique
s you are working that used with the selected technology in your domain. e.g. Identify the distinguished feature and the problem of each
2
Document Page
visualisation technique that used with mixed reality (MR) technology in Surgery guidance in operation room (OR).
Process Steps
Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1

2

3

4

5

Validation Criteria (Measurement Criteria)

Validate (Measure) the solution in each journal paper as below:

How to measure the technique?. Effectiveness of each work should be quantified. What are the parameters (Dependent variables) you are going
to use in assessing you the techniques?; e.g. accuracy, tissue classification, depth perception, speed … etc.

e.g. the parameters to assess the visualisation technique in our samples are accuracy and processing time (speed). You need to be very specific when
you give the evaluation parameters, e.g. you should be very specific which accuracy you measure, registration accuracy, tracking accuracy,

visualisation accuracy … etc.

When you identify the 'Dependent Variables' in each work, you need to identify the parameters (independent variables) that will measure each
dependent variable. e.g. what are the parameters in each work to measure accuracy.

Dependent Variable
Independent Variable
3
Document Page
Input and Output Critical Thinking: Feature of this work,
and Why (Justify)

Critical Thinking:
Limitations of the
research
current solution, and Why
(Justify)

Input (
Data) Output (View)
Mainly: What is the type
of data that should be used

for each solution?. e.g.

Mainly o
Dimensionality:
1D, 2D, 3D, 4D.

Specific: What is the type
of data that is related to

above main that will be

used in each solution? .

e.g.
Raw imaging data ;
acquisition sensor, (e.g.

CT, MRI, fMRI),
Visually
processed data
; semantic,
(e.g. Strategic,

operational),
Derived data;
underlying process, (e.g.

tumour volume

measurement),
Analysed
imaging data
; data
primitive, (e.g. points, line,

contour)

What will be the output
after processing the input

by the techniques?.

Mainly: How it should be
the best be displayed and

interacted with?, e.g.

Mixed reality
; (augmented
reality, virtual reality)

Specific: How it should be
the best be displayed and

interacted with?, e.g.

Perception Location
;
Location, (e.g. Patient,

surgical tool, real

environment, display

device),
Display; Device,
(e.g. 2D technology, laser,

microscope, colour

glasses),
Interaction tool;
Hardware tools, (e.g.

Keyboard, surgical tool)

<
Analyze the solution and find What is the
good features in this work based on your

goal, , and WHY got this good feature.

Based on Critical thinking, each paper you

are working on it, compare it with other you

reviewed and provide YOUR OPINION if

it is better on not than others you reviewed.

You also MUST to JUSTIFY why it is

better or why not ... always link with your

goal when you analyze>

<
Analyze the solution and find what
is/are the limitation/s of it, and WHY

got this limitation >

e.g.
the noise level is still unacceptably
high as the patient berating during the

surgery hasn't considered by the

authored to be filtered.

(
Describe the research/current solution) Evaluation Criteria How this research/current solution is
valuable for your project

<
Describe the research by showing what the author has done to <Validate the major entities of each system. <Link the 'Describe research &
4
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
bring good feature in this work . What is the author contribution ;
how his solution is different from the current solution in this paper

>

e.g.
Gu et al. (2014) investigated the impact of noise on the clarity
of pictures in facial recognition procedures. They offer a solution to

the problem by combining two algorithms that are energy

functionality with deep learning, leading to 78% clarity

improvement.
Place the reference citation
Prove the usefulness and the value of the

system in your project domain for the end

user. >

e.g. the end user in our sample is the

surgeon, to evaluate one of the system

based on the usefulness for the surgeon:

The system help the surgeon to understand

the spatial relationship between anatomical

structure better

limitation sections' in writing you the

how this solution is valuable for the

project>.

e.g.
Whilst the accuracy improvement
over the current solution of 72%
, the
noise level is still unacceptably high
. As
a
result, the combination of algorithms
does
not bring forth possibilities for
refinement

Diagram/Flowchart

<Place all diagrams, flowchart that are related to proposed work>

Version 1.0 _ Week 1 (5 Journal Papers from CSU Library)

5
Document Page
1
Reference in APA format that will be in

'Reference List'

Bini, S. A. (2018). Artificial Intelligence, Machine Learning, Deep Learning, and Cognitive Computing: What Do These

Terms Mean and How Will They Impact Health Care?.
The Journal of arthroplasty.
Citation that will be in the content
Bini, 2018
URL of the
Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
https://www-sciencedirect-

com.ezproxy.csu.edu.au/science/article/

pii/S0883540318302158

Q1
Artificial Intelligence
Machine Learning

Deep Learning

Cognitive Computing

Digital orthopaedics

Digital health

The Name of the Current Solution

(Technique/ Method/ Scheme/

Algorithm/ Model/ Tool/ Framework/ ...

etc )

The Goal (Objective) of this Solution &

What is the Problem that need to be solved

What are the components of it?

Technique/Algorithm name:

Deep learning, Machine learning

Tools:

Biomimicry, Artificial Neural Networks

Applied Area:

Optimizing surgical schedules and

processes, Medical decisions.

Problem:

To identify the method of how the AI will help

in supporting the physicians to deal with the

increasing complexity of the patients’ cases

that they deal with.

Goal:

To tell the readers about the concepts of AI

and the significance its various applications

possess in the health care world today.

AI washing

Pattern recognition

Multiple-layers creation

Self-recoding of the system.

The Process (Mechanism) of this Work; The process steps of the Technique/system

6
Document Page
Process Steps Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1
Giving the inputs:
The AI takes into the inputs given by the

datasets created by previous processes.

This serves as a briefing for the AI about a

number of methods or actions that are taken up

in order to deal with that process.

Insufficient or incorrect data may lead to the

AI resulting in bad initial results until it

learns from its own results later.

2
Analysing:
The results obtained from the corresponding

inputs are analysed by the AI.

This gives an insight of the best probable

solutions over the different possible actions

that could be undertaken.

Incorrect data might mislead the AI into

making up wrong analysis and thereafter

suggesting in the wrong results.

3
Determination:
The best method is identified from studying the

results obtained from the various inputs.

From its analysis of the millions of actions, AI

could suggest the actions that provide the best

results for the particular case.

In cases involving factors other than the

input data, the AI would not be able to

consider them for making a proper

determination for the situation.

Validation Criteria (Measurement Criteria)

Dependent Variable
Independent Variable
Big data
Bigger the size of the dataset, the more is the software has to learn to
become more accurate.

Artificial Neural Networks
Number of layers of the neural networks to be used for the processing of the
data.

Input and Output
Critical Thinking: Feature of this work, and
Why (Justify)

Critical Thinking:
Limitations of the research
current solution, and Why (Justify)

Input (Data)
Output (View) AI uses biomimicry in order to identify the best
health practices and processes because it is able to

The AI is a software which has to fed in by the

datasets in order to enable it to process the data

7
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
The datasets of the
‘big data’ need to be

inputted to the

software so as to

analyse and determine

the actions yielding

the best results.

Through the analysis

of the large datasets,

the AI is able to

identify the actions

that would result into

the best outcomes.

consider all the different factors that are to be

considered while diagnosing a patient, many of

which are often forgotten by human.

and analyse the effectiveness of the actions that

are taken to encounter a problem. If the data is not

sufficient enough or is incorrect or if it lacks some

of the factors associated with the problem, it may

not enable the AI to determine the best solutions.

(
Describe the research/current solution) Evaluation Criteria How this research/current solution is valuable
for your project

The research is aimed at using AI to optimize the

processes of health industry. It mentions the use

of techniques like deep learning and machine

learning by the help of big data and ANNs in

order to come to the best possible solutions for

medical cases that are increasing in criticality

day-by-day.

The AI helps the surgeon or the medical personnel

to attain better results by providing to them many of

the results of some course of actions. It helps to

choose the best probable solution to be taken up.

The research describes how AI has been helpful in

a lot of practical industrial applications. It gives an

outlook over how we can achieve great heights in

healthcare by working in synchronization of man

and machine.

Diagram/Flowchart

8
Document Page
Fig: Complementary support given to humans by AI in making the decisions
2

Reference in APA format that will be in

'Reference List'

Krittanawong, C., Zhang, H., Wang, Z., Aydar, M., & Kitai, T. (2017). Artificial intelligence in precision cardiovascular

medicine.
Journal of the American College of Cardiology, 69(21), 2657-2664.
Citation that will be in the content
Krittanawong et. al., 2017
9
Document Page
URL of the Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
https://www-sciencedirect-

com.ezproxy.csu.edu.au/science/article/

pii/S0735109717368456

Q1
Big Data
Cognitive computing

Deep learning

Machine learning

The Name of the Current Solution

(Technique/ Method/ Scheme/

Algorithm/ Model/ Tool/ Framework/ ...

etc )

The Goal (Objective) of this Solution &

What is the Problem that need to be solved

What are the components of it?

Technique/Algorithm name:

Hypothesis generation, Deep learning

Tools:

Big data, Artificial Neural Networks

Applied Area:

Automated clinical decision systems

Problem:

Cardiovascular clinical care currently has

troubles in tackling the factors like inefficient

care, cost-effectiveness, inadequate patient

care, mortality rates, etc.

Goal:

Through the use of AI, the goal is to provide

the means of better care of patients and

optimum utilization of the medical resources.

It aims at using them for cardiovascular

disease diagnosis and predictions.

Supervised Learning

Unsupervised Learning

Reinforced Learning

Deep Learning

Cognitive Computing

The Process (Mechanism) of this Work; The process steps of the Technique/system

Process Steps
Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1
Big Data: It may help to identify the factors for a It is tedious process where the dataset has to
10
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
Processing of big data to make predictions. problem and the various results obtained from
the different kinds of diagnosis done over it in

the past.

be formed at first in order to let the software

process and perform its tasks.

2
Supervised learning:
Training a machine to perform according to the

datasets that are fed into the software.

It helps in prediction, diagnosis and the

treatment of diseases.

Small training datasets or biased datasets

may result in the inaccurate decisions.

3
Unsupervised learning:
Determination of the hidden patterns in data by

the help of algorithms like clustering algorithm

and association rule-learning algorithm.

The unknown results and patterns are

identified by the use of unsupervised learning.

Determination of the initial cluster pattern is

very important. If done wrongly, it could

lead to biased final results.

4
Deep learning:
Use of deep learning algorithms in order to

facilitate the CV imaging in order to generate

the automated predictions from the data that has

been inputted to the software.

It mimics how the human brain works by

creating a number of layers of neural networks,

thus unfolding many hidden patterns.

Nonlinear analysis or overfitting (too many

parameters compared to the dataset) may

lead to inaccurate predictions.

5
Cognitive computing:
Use of self-learning systems to mimic the

human thinking by the use of machine learning,

pattern recognition, and natural learning process.

It could be used to solve the problems or cure

the diseases without any human interference at

all.

There is a fear of initiation of the AI era.

Validation Criteria (Measurement Criteria)

Dependent Variable
Independent Variable
Learning curves
Area under the curve
c-statistics
Sample size
Pattern recognitions
Over-fitting and under-fitting of the data
11
Document Page
Input and Output Critical Thinking: Feature of this work, and
Why (Justify)

Critical Thinking:
Limitations of the research
current solution, and Why (Justify)

Input (Data)
Output (View)
Big data and

algorithms.

Hidden patterns and

solutions yielded from

the algorithms.

This article describes the role of various AI

technologies that are helpful in treating the patients

more effectively. These will help in predicting the

diseases and also in curing them with the best

possible cure.

The research does not describe how the cognitive

computing would be different from the other

techniques.

(
Describe the research/current solution) Evaluation Criteria How this research/current solution is valuable
for your project

Big data and algorithms when used in the AI

software along with its machine learning and

cognitive computing helps the physicians in

predicting, diagnosing and curing their patients

by the solutions they receive from the AI.

The results like decrease in the costs and increase in

the patients’ care provide ample criteria to analyse

the effectiveness of the AI solutions for the

healthcare industry.

This research shares so many examples of CV

patients getting help by the AI which is a mark-up

for the project.

Diagram/Flowchart

12
chevron_up_icon
1 out of 80
circle_padding
hide_on_mobile
zoom_out_icon
[object Object]