A PROJECT REPORT ON

ALZHEIMER’S DISEASE DETECTION USING MACHINE LEARNING

TECHNIQUES IN 3D MR IMAGES

SUBMITTED TO THE SAVITRIBAI PHULE PUNE UNIVERSITY, PUNE

IN THE PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE AWARD OF THE DEGREE
OF

BACHELOR OF ENGINEERING
(COMPUTER ENGINEERING)

SUBMITTED BY

Hemangi Akhade B190614207

Aishwarya Bharti B190614225
Divya Patil B190614242
Vaishnavi Shelke B190614323

UNDER THE GUIDANCE OF

Prof. Pramod G Patil

DEPARTMENT OF COMPUTER ENGINEERING

Sandip Institute of Technology and Research Centre, Nashik

Nashik-422213

SAVITRIBAI PHULE PUNE UNIVERSITY

2022 -2023
 CERTIFICATE
This is to certify that the project report entitles

“ALZHEIMER’S DISEASE DETECTION USING MACHINE LEARNING

TECHNIQUES IN 3D MR IMAGES”

Submitted by

Hemangi Akhade B190614207

Aishwarya Bharti B190614225
Divya Patil B190614242
Vaishnavi Shelke B190614323

is a bonafide student at this institute and the work has been carried out by him/her under
the supervision of Prof. Pramod G. Patil and it is approved for the partial fulfillment of
the requirement of Savitribai Phule Pune University, for the award of the degree of
Bachelor of Engineering (Computer Engineering).

Prof. Pramod G Patil Dr. Amol Potgantwar

Guide Head
Department of Computer Engineering Department of Computer Engineering

External Examiner Dr. M. M. Patil

Principal

DEPARTMENT OF COMPUTER ENGINEERING

Sandip Institute of Technology and Research Centre, Nashik
Nashik-422213

SAVITRIBAI PHULE PUNE UNIVERSITY

2022 -2023
 ACKNOWLEDGEMENT

It gives us great pleasure in presenting the project report on the topic ‘Alzheimer’s
Disease Detection Using Machine Learning Techniques In 3D MR Images’.

We would like to express our heartfelt gratitude to our internal guide, Prof. Pramod G.
Patil, for providing us with the necessary help and guidance we needed. We are truly
grateful for their kind support and valuable suggestions.

We are also thankful to Dr. Amol D. Potgantwar, Head of the Computer Engineering
Department at SITRC, for his invaluable support and suggestions.

Furthermore, we extend our gratitude to Dr. M. M. Patil for providing various resources
such as a well-equipped laboratory with the required software platforms and a continuous
internet connection for our project.

Lastly, I would like to extend my sincere thanks to our project coordinator, Prof. Sunil
Kale, for his constant support and guidance throughout the completion of this project.

Hemangi Akhade B190614207

Aishwarya Bharti B190614225
Divya Patil B190614242
Vaishnavi Shelke B190614323
 ABSTRACT

Alzheimer's disease, a progressive neurodegenerative disorder impacting memory,

cognition, and behavior, necessitates early identification for effective intervention and
treatment. This study evaluates the performance of a convolutional neural network
(CNN) model for Alzheimer's disease (AD) classification based on MRI image
processing. This study proposes a deep learning framework that utilizes magnetic
resonance imaging (MRI) scans to detect Alzheimer's disease. The potential of the CNN
model in accurately identifying AD cases using MRI scans, emphasizes its effectiveness
as a diagnostic tool for early detection and intervention in AD. Specifically, a
convolutional neural network (CNN) is trained on a substantial dataset of MRI scans,
enabling the automatic identification of distinctive patterns between healthy individuals
and those with Alzheimer's disease. The performance of this approach is assessed on an
independent test set, yielding promising. By enhancing the accuracy and speed of
Alzheimer's disease detection, our proposed approach holds the potential to facilitate
earlier intervention and improve patient outcomes.

Keywords: Alzheimer's disease, Convolutional Neural Network, Magnetic Resonance

Imaging, Deep learning, Machine learning, Image processing.
 TABLE OF CONTENTS

LIST OF ABBREVATIONS i
LIST OF FIGURES ii
LIST OF TABLES iii

Sr. No. TITLE OF CHAPTER PAGE NO.

01 Introduction 1-4
1.1 Overview
1.2 Motivation
1.3 Problem Definition
02 Literature Survey 5-14
03 Software Requirements Specification 15-26
3.1 Introduction 16
3.1.1 Project Scope
3.1.2 User Classes and Characteristics
3.1.3 Assumptions and Dependencies
3.2 Functional Requirements 18
3.3 External Interface Requirements 18
3.3.1 User Interfaces
3.3.2 Hardware Interfaces
3.3.3 Software Interfaces
3.3.4 Communication Interfaces
3.4 Nonfunctional Requirements 19
3.4.1 Performance Requirements
3.4.2 Safety Requirements
3.4.3 Software Quality Attributes
3.5 System Requirements 20
3.3.1 Database Requirements
3.3.2 Software Requirements
5.3.3 Hardware Requirements
3.6 Analysis Models: SDLC Model to be applied 21
3.7 System Implementation Plan 24
04 System Design 27-35
4.1 System Architecture
4.2 Mathematical Model
4.3 Data Flow Diagrams
4.4 UML Diagrams
05 Other Specification 36-37
 5.1 Advantages
5.2 Limitations
5.3 Applications
06 Project Implementation 38-40
6.1 Algorithm & Flowcharts
07 Results & Screenshots 41-46
08 Conclusion & Future Work 47-48
09 References 49-53
10 Information of Project Group Members 54-58
11 Research Paper 59-61
12 Paper Publication Certificate 62-67
13 Sponsorship Letter 68-70
14 Plagiarism Report 71-72
 LIST OF ABBREVATIONS

ABBREVIATION ILLUSTRATION
AD Alzheimer’s Disease
CNN Convolutional Neural Networks
DFD Data Flow Diagram
MRI Magnetic Resonance Image
UML Unified Modeling Language
 LIST OF FIGURES

Figure ILLUSTRATION PAGE NO.

Figure 3.1 Waterfall Model 22
Figure 4.1 System Architecture 28
Figure 4.1 Data Flow(0) Diagram 30
Figure 4.2 Data Flow(1) Diagram 30
Figure 4.3 Data Flow(2) Diagram 31
Figure 4.4 Class Diagram 32
Figure 4.5 Use Case Diagram 33
Figure 4.6 Activity Diagram 34
Figure 4.7 Sequence Diagram 35
Figure 6.1 Flowchart 39
 LIST OF TABLES

Table Illustration PAGE NO.

Table 3. 1 Project Planning Chart 25
Table 3. 2 Task Scheduling 26
 CHAPTER 1

INTRODUCTION

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 1

1.1 Overview

Alzheimer's Disease (AD) is a neurodegenerative disorder that primarily affects the

elderly population. It is characterized by progressive cognitive decline, memory loss, and
impaired daily functioning. Unfortunately, there is currently no known treatment to halt
or reverse the progression of the disease. The prevalence of AD has been steadily
increasing, and projections indicate a significant rise in the number of affected
individuals in the coming years. Reports from 2005 through 2030 suggest a continuous
growth in the percentage estimate of people affected by AD. Currently, approximately 40
million individuals worldwide suffer from AD, and it is projected that this number could
reach 135 million by 2050. In recent years, advancements in medical imaging
technology, particularly Magnetic Resonance Imaging (MRI), have enabled researchers
and healthcare professionals to study the structural and functional changes that occur in
the brains of AD patients. MRI scans provide detailed images of the brain, allowing for
the detection of specific biomarkers and abnormalities associated with AD. Additionally,
the integration of machine learning techniques, such as Convolutional Neural Networks
(CNNs), has shown promise in automating the analysis and interpretation of MRI scans
for AD detection. This project aims to leverage the power of CNNs and MRI scans to
develop a robust and accurate algorithm for the early detection of AD. By training a CNN
model on a diverse dataset of MRI scans, we seek to enable healthcare professionals to
identify potential cases of AD more efficiently and effectively. The ultimate goal is to
provide a tool that can assist in screening individuals for AD, facilitating early
interventions, treatment planning, and improved management of the disease.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 2

1.2 Motivation

The main motivation of our project is to develop a system that can accurately detect
Alzheimer's disease using advanced technologies like Convolutional Neural Networks
(CNN) and Magnetic Resonance Imaging (MRI). Our goal is to catch the disease at an
early stage so that doctors can provide timely care and treatment to patients. By using
CNN and MRI, we aim to minimize incorrect diagnoses, unnecessary tests, and improve
patient outcomes.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 3

1.3 Problem Definition

The problem we aim to address is the early detection of Alzheimer's disease using MRI
scans. The current methods are slow, costly, and may not catch early signs. Our idea is to
use advanced computer technology called deep learning to quickly analyze MRI scans
and find patterns that suggest Alzheimer's disease. This can help us detect the disease
more accurately and quickly, leading to better treatment and outcomes for patients.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 4

 CHAPTER 2

LITERATURE SURVEY

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 5

1. Paper Name: Use of Non-linear and Complexity Features for EEG-Based Dementia
Alzheimer Disease Diagnosis.
Author: Nilesh. N. Kulkarni, Saurabh. V. Parhad, Yasmin. P. Shaikh.
Abstract: Alzheimer's disease is one of the most common and fastest-growing
neurodegenerative diseases in Western countries. The development of different
biomarkers tools is a key issue for the diagnosis of Alzheimer's disease and its
progression. Prediction of the cognitive performance of subjects from EEG and
identification of relevant biomarkers are some of the research problems. EEG signal
analysis can be well suited for automated diagnosis of Alzheimer’s disease. Although
EEG-based techniques help screen Alzheimer's and dementia, still there is a scope for
improvement in terms of diagnostic accuracy, sensitivity, and specificity. Thus, many
issues are still left out in the field of Alzheimer's diagnosis using EEG signals related to
the choice of features that can help in distinguishing two or more subjects. This paper
focuses on new features for the diagnosis of Alzheimer’s disease using EEG signals with
an effective increase in diagnostic accuracy. The use of new complexity-based features is
proposed in this paper which increases diagnostic accuracy and helps in early
Alzheimer’s diagnosis.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 6

2. Paper Name: Multivariate Analysis of Structural MRI and PET (FDG and 18FAV-45)
for Alzheimer’s Disease and Its Prodromal Stages.
Author: Qi Zhou, Mohammed Goryawala, Mercedes Cabrerizo, Warren Barker, David
Loewenstein, Ranjan Duara and Malek Adjouadi.
Abstract: A multivariate analysis method, orthogonal partial least squares to latent
structures (OPLS), was used to discriminate Alzheimer’s disease (AD), early and late
mild cognitive impairment (EMCI and LMCI) from cognitively normal control (CN)
using MRI and PET measures. Free Surfer 5.1 generated 271 MRI features including 49
subcortical volumes, 68 cortical volumes, 68 cortical thicknesses, 70 surface areas and 16
hippocampus subfields. Subjects with all aforementioned MRI measures passing quality
control and valid Fludeoxyglucose (18F) (FDG) and Florbetapir (18F) PET scans were
selected from ADNI database, resulting in a total of 524 participants (137 CN, 214
EMCI, 103 LMCI and 70 AD) for the study. Altogether 286 features including 15
significant PET uptake features (7 for FDG and 8 for AV-45) were utilized for OPLS
analysis. Predictive power was evaluated by ( ), a quantifier of the statistical significance
for class separation. The results show that MRI features ( ( ) =0.645), and PET features ( (
) = 0.636) has comparable predictive power in separating AD from CN, and MRI features
are better predictor of LMCI ( ( ) = 0.282) than PET ( ( ) = 0.294). Combination of PET
and MRI has the most predictive power for LMCI and AD with ( ) of 0.294 and 0.721,
respectively. While for EMCI, cortical thickness was found to be the best predictor with a
( ) of 0.108, suggesting cortical thickness may be the first structural change ahead of
others and should be prioritized in prediction of very mild cognitive impairment.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 7

3. Paper Name: A Novel Gene Selection Method using GA/SVM and Fisher Criteria in
Alzheimer’s Disease.
Author: Seyede Zahra Paylakhi, Sadjaad Ozgoli, Seyed Hassan Paylakhi.
Abstract: Identification of those genes which cause diseases can develop the process of
diagnosis and the treatment of diseases. In this paper, a gene selection method based on
genetic algorithm (GA) and support vector machines (SVM) is presented. At first, Fisher
criteria is utilized in order to do filtration for those genes which are noisy and redundant
in high dimensional micro array data. Then, GA/SVM model is used for selection of
various subsets of maximally informative genes with the use of different training sets.
The frequency of appearance of each gene in various subsets of genes is analyzed.
Therefore, the last subset contains those genes which are highly informative. In fact,
Fisher and GA/SVM methods have been merged in order to take benefit from a filtering
method as well as an embedded method. The proposed method is tested on DNA
microarray gene expression data of Alzheimer’s disease. The results show that the
proposed method has a good selection and classification performance, which can yield
100 biological point of view, at least 8 ( 53 Alzheimer associated genes. Thus, these
genes not only can serve as predictors of the disease, but also can use as a means to find
new candidate.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 8

4. Paper Name: A Novel Gaussian Discriminant Analysis-based Computer Aided
Diagnosis System for Screening Different Stages of Alzheimer’s Disease.
Author: Chen Fang, Chunfei Li, Mercedes Cabrerizo, Armando Barreto, Jean Andrian,
David Loewenstein, Ranjan Duara, Malek Adjouadi.
Abstract: This study introduces a novel Gaussian discriminant analysis (GDA)- based
computer aided diagnosis (CAD) system using structural magnetic resonance imaging
(MRI) data uniquely as input for screening different stages of Alzheimer’s disease (AD)
involving its prodromal stage of mild cognitive impairment (MCI) in relation to the
cognitive normal control group (CN). Taking advantage of multiple modalities of
biomarkers, over the past few years, several machine learning based CAD approaches
have been proposed to address this high-dimensional classification problem. This study
presents a novel GDA-based CAD system on the basis of a tenfold cross validation and a
held-out test data set. Subjects considered in this study included 187 CN, 301 MCI, and
131 AD subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI)
database. In the tenfold cross validation, the proposed system achieved an average F1
score of 97.20 sensitivity of 99.14 discriminating together the MCI and AD groups from
the CN group; and an average F1 score of 79.8287.43 discriminating AD from MCI. By
testing on the held-out test data, for discriminating MCI and AD from CN, an accuracy of
93.28 were obtained. These results also show that by separating left and right
hemispheres of the brain into two decisional spaces, and then combining their outputs,
the GDA-based CAD system demonstrates a high potential for clinical application.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 9

5. Paper Name: Comparison Analysis of Machine Learning Algorithms to Rank
Alzheimer’s Disease Risk Factors by Importance.
Author: Mohamed Mahyoub, Dr. Martin Randles, Dr. Thar Baker, Dr Po Yang.
Abstract: People have always feared aging, and the increasing rate of dementia disease
caused this fear to twofold. Dementia is irreversible, unstoppable and has no known cure.
According to Alzheimer’s Disease International 2015 and World Alzheimer Report 2015,
the estimated financial cost for healthcare services of Alzheimer’s Disease is Trillion in
2018. This paper discusses the both behavioural and biological markers data, and a
computational method to rank Alzheimer’s Disease risk factors by importance using
different machine learning models on Alzheimer’s Disease clinical assessment data from
ADNI. The dataset contains Alzheimer’s Disease risk factors data related to medical
history, family dementia history, demographical, and some lifestyle data for 1635
subjects. There are 387 normal control, 87 significant memory concerns, 289 early mild
cognitive impairment, 539 late mild cognitive impairment and 333 Alzheimer’s Disease
subjects. We deployed different machine learning models on the dataset to rank the
importance of the variables ( risk importance of investigating Alzheimer’s Disease using
machine learning, the need to use factors ). The results show that some risk factors in
subjects genetically, demography and lifestyle are more important than some medical
history risk factors. Having APOE4, education level, age, weight, family dementia
history, and type of work rank as more influential among Alzheimer’s Disease subjects.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 10

6. Paper Name: Segmentation of Tau Stained Alzheimer’s Brain Tissue using
Convolutional Neural Networks.
Author: Alexander Wurts, Derek H. Oakley, Bradley T. Hyman, Siddharth Samsi.
Abstract: Alzheimer’s disease is characterized by complex changes in brain tissue
including the accumulation of tau-containing neurofibrillary tangles (NFTs) and
dystrophic neurites (DNs) within neurons. The distribution and density of tau pathology
throughout the brain is evaluated at autopsy as one component of Alzheimer’s disease
diagnosis. Deep neural networks (DNN) have been shown to be effective in the
quantification of tau pathology when trained on fully annotated images. In this paper, we
examine the effectiveness of three DNNs for the segmentation of tau pathology when
trained on noisily labeled data. We train FCN, SegNet and U-Net on the same set of
training images. Our results show that using noisily labeled data, these networks are
capable of segmenting tau pathology as well as nuclei in as few as 40 training epochs
with varying degrees of success. SegNet, FCN and U-Net are able to achieve a DICE loss
of 0.234, 0.297 and 0.272 respectively on the task of segmenting regions of tau. We also
apply these networks to the task of segmenting whole slide images of tissue sections and
discuss their practical applicability for processing gigapixel sized images.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 11

7. Paper Name: Radiomics textural features extracted from subcortical structures of grey
matter probability for Alzheimer’s disease detection.
Author name: Cesar A. Ortiz Toro, Nuria Gutierrez Sanchez, Consuelo Gonzalo-Martin,
Roberto Garrido Garcia, Alejandro Rodriguez Gonzalez and Ernestina Menasalvas Ruiz.
Abstract: Alzheimer’s disease (AD) is characterized by a progressive deterioration of
cognitive and behavioral functions as a result of the atrophy of specific regions of the
brain. It is estimated that by 2050 there will be 131.5 million people affected. Thus, there
is an urgent need to find biological markers for its early detection and monitoring. In this
work, it is present an analysis of textural radiomics features extracted from a gray matter
probability volume, in a set of individual subcortical regions, from a number of different
atlases, to identify subject with AD in a MRI. Also, significant subcortical regions for
AD detection have been identified using a Relief relevance test. Experimental results
using the ADNI1 database have proven the potential of some of the tested radiomic
features as possible biomarkers for AD/CN differentiation.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 12

8. Paper Name: A Neuroimaging Feature Extraction Model for Imaging Genetics with
Application to Alzheimer’s Disease.
Author: Chunfei Li, Chen Fang, Malek Adjouadi, Mercedes Cabrerizo, Armando
Barreto, Jean Andrian, Ranjan Duara, David Loewenstein.
Abstract: Neuroimaging is an important research platform that can be very useful for
eliciting new understanding on the complicated pathogenesis between genetics and
disease phenotypes. Due to the extremely high dimensionality of image and genetic data,
and considering the potential joint effect of genetic variants, multivariate techniques have
been examined to detect Alzheimer’s disease (AD) related genetic variants expressed
through single-nucleotide polymorphisms (SNPs). However, the image features used in
support of those methods are not immediately related to the disease, and the detected
genetic markers may not be related to AD. In this study, we propose an ensemble model
based framework for firstly extracting 50 region based image features whose values are
predicted by base learners trained on raw neuroimaging morphological variables. This
task is followed by performing sparse Partial Least Squares regression (sPLS) method on
the extracted 50 AD related image features and pre-selected 1508 SNPs to detect the
significant SNPs associated with the extracted image features. Instead of modeling a
direct link between genetic variants and disease label, we captured disease information
indirectly.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 13

9. Paper name: Automatic Recognition of Mild Cognitive Impairment and Alzheimer’s
Disease Using Ensemble based 3D Densely Connected Convolutional Networks
Author: Shuqiang Wang, Hongfei Wang, Yanyan Shen, Xiangyu Wang.
Abstract:—Automatic diagnosis of Alzheimer’s disease (AD) and mild cognition
impairment (MCI) from 3D brain magnetic resonance (MR) images plays an important
role in early treatment of dementia disease. Deep learning architectures can extract
potential features of dementia disease and capture brain anatomical changes from MRI
scans. This paper proposes an ensemble of 3D densely connected convolutional networks
(3D-DenseNets) for AD and MCI diagnosis. First, dense connections were introduced to
maximize the information flow, where each layer connects with all subsequent layers
directly. Then weighted-based fusion method was employed to combine 3D-DenseNets
with different architectures. Extensive experiments were conducted to analyze the
performance of 3D-DenseNet with different hyper-parameters and architectures. Superior
performance of the proposed model was demonstrated on ADNI dataset including 833
subjects.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 14

 CHAPTER 3

SOFTWARE REQUIREMENTS
SPECIFICATION

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 15

3.1 Introduction

3.1.1 Project Scope

The goal of this project is to develop a system that utilizes CNN and MRI technology to
detect Alzheimer's disease at an early stage. The system will involve collecting MRI
scans from individuals with and without Alzheimer's disease, preprocessing the data, and
designing a CNN model. The trained model will be capable of accurately identifying
Alzheimer's disease based on the MRI scans.

3.1.2 User Classes and Characteristics

1. Login/Registration:

The system includes a login and registration feature, allowing users to access their
accounts and obtain results. Users have the option to log in if they already have an
account or register if they are new users. This feature ensures that only authorized
individuals can utilize the system and view the results.

2. User and Administrator Proficiency:

It is assumed that users possess basic knowledge of operating the internet and have access
to an internet connection. Users should be familiar with common internet browsing
techniques and be able to navigate through the system's interface. Additionally, the
administrator is expected to be familiar with the interface of the tech support system.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 16

3.1.3 Assumption and Dependencies Platforms

1. Python Language: The project relies on the use of Python as the primary programming
language for implementing machine learning (ML) algorithms. Python offers a wide
range of libraries and frameworks specifically designed for ML tasks, making it a popular
choice in the field.

2. Machine Learning: The project relies on ML techniques to develop the Alzheimer's

disease detection model. ML algorithms, such as Convolutional Neural Networks
(CNNs), will be employed to analyze the input image data and make predictions.

3. Input as Image through Image Processing: The project assumes that the input data for
the model will be in the form of images. These images will undergo various image
processing techniques to extract relevant features required for accurate Alzheimer's
disease detection.

4. Software: The project utilizes the Anaconda Navigator software, which provides an
integrated development environment (IDE) and package management for Python.
Specifically, the IDE Spyder is used for coding and developing the ML model.

5. Libraries: The project relies on several Python libraries for implementing various
functionalities. The main libraries used include Keras (for building and training deep
learning models), OpenCV (for image processing tasks), TensorFlow (for ML
computations), Matplotlib (for data visualization), and Pillow (for image manipulation).

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 17

3.2 Functional Requirements

1. Understanding the problem statement.

2. Identifying the hardware and software requirements of the proposed system.
3. Gaining a comprehensive understanding of the proposed system.
4. Planning various activities with the assistance of a planner.
5. Performing tasks such as designing, programming, and testing.

3.3 External Interface Requirements

3.3.1 User Interfaces

Application-based Alzheimer’s disease detection.

3.3.2 Hardware Interfaces

• System : Intel I5 Processor

• Hard Disk : 40 GB
• Monitor : 15
• RAM : 8 GB

3.3.3 Software Interfaces

• Operating system : Windows 10

• Coding Language : Python
• IDE : Spyder
• Database : SQLite

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 18

3.3.4 Communication Interfaces

The system can work only in online mode hence, communication interfaces are
compulsory.

3.4 Non-Functional Requirements

3.4.1 Performance Requirements

1. The performance of the functions and every module must be well.

2. The overall performance of the software will enable the users to work efficiently.
3. Performance of the response should be fast.
4. Performance of the providing virtual environment should be fast.

3.4.2 Safety Requirements

1. The application is designed in modules where errors can be detected and fixed easily.
2. This makes it easier to install and update new functionality if required.

3.4.3 Software Quality Attributes

Our software has many quality attributes that are given below:
1. Adaptability: This software is adaptable by all users.
2. Availability: This software is freely available to all users. The availability of the
software is easy for everyone.
3. Maintainability: After the deployment of the project if any error occurs then it can be
easily maintained by the software developer.
3. Reliability: The performance of the software is better which will increase the reliability
of the Software.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 19

4. User Friendliness: Since the software is a GUI application; the output generated is
much user friendly in its behavior.
5. Integrity: Integrity refers to the extent to which access to software or data by
unauthorized persons can be controlled.
6. Security: Users are authenticated using many security phases so reliable security is
provided.
7. Testability: The software will be tested considering all aspects.

3.5 System Requirements

3.5.1 Database Requirement

• Database : DBSQLite

3.5.2 Software Requirements

• Operating system : Windows 10

• Coding Language : Python
• Software : Anaconda Navigator
• IDE : Spyder
• Database : DBSQLITE
• Front-end : TKinter ( GUI )
• Back-end : Python

3.5.3 Hardware Requirements

• System : Intel I5 Processor

• Hard Disk : 40 GB
• Monitor : 15
• Ram : 8 GB

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 20

3.6 Analysis Models: SDLC Model to be applied

SDLC - Waterfall Model

The first Process Model to be introduced was the Waterfall Model. The term "linear
sequential life cycle model" is also used to describe it. It is incredibly easy to use and
comprehend. There is no overlap between phases in a waterfall model; each step must be
finished before the subsequent phase can start. The first SDLC methodology for software
development was the waterfall model. The software development process is depicted
using the waterfall model, which follows a linear sequential flow. This implies that a
phase of development can only start if the one before it is finished. The phases in this
waterfall model do not cross over.

Waterfall Model – Design

The Waterfall Approach was the first commonly used SDLC Model in software
engineering to ensure project success. The entire software development process is split
into distinct phases using "The Waterfall" technique. Typically, the results of one step in
this waterfall model serve as the input for the subsequent phases in turn. The several
stages of the Waterfall Model are depicted in the figure.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 21

 Figure 3.1 Waterfall Model

The sequential phases in the waterfall model are -

• Requirement Analysis and Gathering - During this stage, all prospective system
requirements are gathered and compiled into a requirement specification document.

• System Design - In this phase, the required specifications from the first phase are
examined, and the system design is created. This system design aids in determining
the overall system architecture as well as the hardware and system requirements.

• Implementation - The system is initially built-in discrete programs known as units,

which are then combined in the following phase, using inputs from the system design.
Unit testing is the process of developing and evaluating each unit for functionality.

• Integration and Testing - Following the testing of each unit created during the
implementation phase, the entire system is merged. The entire system is tested for
errors and failures after integration.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 22

• Deployment of system - Once the functional and non-functional testing is done; the
product is deployed in the customer environment or released into the market.

• Maintenance - Various problems can arise in a client environment. Patches are

published to address certain problems. Additionally, some improved versions of the
product have been launched. To bring about these changes in the surroundings of the
consumer, maintenance is performed. The progression is perceived as flowing slowly
downward (like a waterfall) through the phases as all of these phases are connected.
The "Waterfall Model" gets its name because the following phase doesn't begin until
the prior phases established set of goals has been met and it has been approved. Phases
do not cross over in this model.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 23

3.7 System Implementation Plan

3.7.1 Project Schedule

Sr. Task Description Duration

No.

1 Domain Selection Choosing the domain for the paper. 15

2 Deciding the topic for Choosing a specific topic for the project. 15

the project

3 Paper Selection Selecting relevant papers. 10

4 Requirement Gathering Collecting project requirements. 20

5 Literature Survey Conducting a survey of existing literature. 15

6 Problem Identification Identifying the problem to be addressed. 15

7 Original Designing the initial architecture. 15

Architecture

8 Modified Architecture Modifying the architecture based on 25

requirements.

9 Original Algorithm Developing the initial algorithm. 20

11 Modified Algorithm Modifying the algorithm as needed. 30

12 Mathematical Model Creating a mathematical model for the 30

system.

13 Analysis Analyzing the system's performance and 30

results.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 24

 14 Planning Creating a project plan. 15

15 Design Creating the system design. 30

16 Coding Implementing the system in code. 30

17 Testing Performing tests on the system. 15

Table 3. 1 Project Planning Chart

3.7.2 Timeline Chart

Figure 3.2 Timeline Chart

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 25

3.7.3 Task Scheduling

Task Milestone Name Number of Number of

Weeks Days

Task 1 Project Team Formation 2 14

Task 2 Problem Identification and Understanding 2 14

Task 3 Presentation and Topic Finalization 3 21

Task 4 Study IEEE Paper 3 21

Task 5 Synopsis Submission 2 14

Task 6 Requirement Specification and Analysis of 4 30

Existing Technology

Task 7 Detailed Designing 5 36

Task 8 Implementation Module 1 2 14

Task 9 Implementation Module 2 2 14

Task 10 Implementation Module 3 2 14

Task 11 Implementation Module 4 3 21

Task 12 Integrating Modules and Software Testing 4 30

Task 13 Documentation 4 30

Table 3. 2 Task Scheduling

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 26

 CHAPTER 4

SYSTEM DESIGN

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 27

4.1 System Architecture

Figure 4.1 System Architecture

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 28

4.2 Mathematical Model

• Let S be the Whole system S = { I, P, O }

• I - Input

• P - Procedure

• O - Output

• Input ( I )

• I = { brain Images }

• Where,

• Images -> brain images

• Procedure ( P ),

• P={ I, Using I System perform operations and calculate the detection of disease.}

• Output ( O )

• O = { System detect the Alzheimer's disease or not }

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 29

4.3 Data Flow Diagram

In a Data Flow Diagram (DFD), we depict the flow of data within a system. In DFD0, we
represent the basic DFD where rectangles represent inputs and outputs, and circles
represent the system itself. In DFD1, we illustrate the actual inputs and outputs of the
system. The inputs to our system can be in the form of text or images, and the output is
the detection of rumors. Similarly, in DFD2, we present the operations performed by both
users and administrators.

Figure 4.2 Data Flow(0) Diagram

Figure 4.3 Data Flow(1) Diagram

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 30

 Figure 4.4 Data Flow(2) Diagram

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 31

4.4 UML Diagrams

The Unified Modeling Language (UML) is a standard language for writing software
blueprints. It can be used to visualize, specify, construct, and document the artifacts of a
software-intensive system. UML is process-independent, although it is best suited for use
in a process that is use case driven, architecture-centric, iterative, and incremental. There
are several UML diagrams available that can be utilized to represent different aspects of
the system's structure and behavior.

4.4.1 Class Diagram

Figure 4.5 Class Diagram

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 32

4.4.2 Use Case Diagram

Figure 4.6 Use Case Diagram

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 33

4.4.3 Activity Diagram

Figure 4.7 Activity Diagram

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 34

4.4.4 Sequence Diagram

Figure 4.8 Sequence Diagram

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 35

 CHAPTER 5

OTHER SPECIFICATION

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 36

5.1 Advantages

1. Detects Alzheimer's disease early, leading to timely treatment and better outcomes.
2. Saves time by automating the diagnosis process.
3. Reduces costs by minimizing the need for multiple appointments and tests.
4. Improves efficiency in detecting Alzheimer's disease.
5. Enhances patient care and improves quality of life.

5.2 Limitations

Our model, like other deep learning models, struggles to accurately estimate uncertainty.
It means that the model may have difficulty indicating how confident or unsure it is about
its predictions. However, if we combine statistical techniques with deep learning and use
active learning, we can potentially improve the model's performance in analyzing 3D
brain scans for Alzheimer's detection.

5.3 Applications

The system can be used in hospitals to help detect Alzheimer's patients. Early detection
of the disease can assist in receiving proper and timely treatment.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 37

 CHAPTER 6

PROJECT IMPLEMENTATION

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 38

6.1 Algorithm & Flowcharts

Figure 6.1 Flowchart

Algorithm:

1. Gather a dataset of brain images from patients with Alzheimer's disease and healthy
individuals.

2. Clean and preprocess the images, ensuring proper formatting and analysis readiness.
This includes resizing, normalization, and standardization.

3. Split the dataset into a training set and a testing set.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 39

4. Extract features from the brain images using a pre-trained CNN model that can
identify relevant features for Alzheimer's disease prediction.

5. Train a new model using the extracted features as inputs and the labels (Alzheimer's
or healthy) as outputs.

6. Validate the trained model on the testing set to ensure accurate prediction of
Alzheimer's disease in new, unseen brain images.

7. Evaluate the model's performance using various metrics such as accuracy, precision,
recall, and F1 score.

8. Fine-tune the model by adjusting its hyperparameters to optimize performance.

9. Deploy the trained model in a real-world scenario, where it can predict Alzheimer's
disease in patients based on their brain images.

10. Monitor and update the model over time to maintain its accuracy and effectiveness in
predicting Alzheimer's disease.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 40

 CHAPTER 7

RESULTS AND SCREENSHOTS

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 41

7.1 Results & Screenshots

Welcome Page

Sign Up Page

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 42

 Login Page

Home Page

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 43

 Select Image

Image Pre-processing

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 44

 No Alzheimer Detected

Mild Alzheimer Detected

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 45

 Very Mild Alzheimer Detected

Moderate Alzheimer Detected

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 46

 CHAPTER 8

CONCLUSION AND FUTURE WORK

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 47

6.1. Conclusion

In conclusion, our project uses a combination of CNN and MRI to detect Alzheimer's
disease. We achieved a high accuracy rate of 85% to 90% in identifying the disease from
MRI scans. This helps doctors diagnose Alzheimer's more accurately and provide timely
care. CNN technology helps us extract important features and classify MRI images
effectively. Our project contributes to a better understanding and management of
Alzheimer's disease, benefiting both patients and healthcare professionals.

6.2. Future Work

In the future, we will focus on improving CNN models by training them on larger and
more diverse datasets. We aim to integrate additional biomarkers and clinical data with
MRI scans for more comprehensive diagnostics.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 48

 CHAPTER 9

REFERENCES

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 49

1. Sarraf S., Tofighi G., DeepAD: Alzheimer's Disease Classification via Deep
Convolutional Neural Networks using MRI and fMRI, BioRxiv, 2016.

2. Suk H., Lee S.W., Shen D., Latent Feature Representation with Stacked Auto-
encoder for AD/MCI Diagnosis, Brain Structure and Function, 2016.

3. Plis S., Hjelm D., Salakhutdinov R., Allen E., Alzheimer's Disease Neuroimaging
Initiative, Deep Learning for Neuroimaging: A Validation Study, Frontiers in
Neuroscience, 2014.

4. Popuri K., Cobzas D., Alzheimer's Disease Neuroimaging Initiative, et al., A

variational approach for classification of AD using brain imaging data, IEEE
Transactions on Medical Imaging, 2011.

5. Zhang D., Shen D., Multi-modal multi-task learning for joint prediction of multiple
regression and classification variables in Alzheimer's disease, NeuroImage, 2012.

6. Payan A., Montana G., Predicting Alzheimer's Disease: A Neuroimaging Study

with 3D Convolutional Neural Networks, arXiv preprint, 2015.

7. Liu M., Zhang D., et al., Fusion of Multimodal Neuroimaging Data via Sparse
Representation Based Classification for Alzheimer's Disease Diagnosis, Human
Brain Mapping, 2014.

8. Sarraf S., Ghafoorian M., et al., DeepAD: Alzheimer's Disease Classification via
Deep Convolutional Neural Networks using MRI and fMRI, Scientific Reports,
2019.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 50

9. Suk H., Lee S.W., et al., Hierarchical Feature Representation and Multimodal
Fusion with Deep Learning for AD/MCI Diagnosis, NeuroImage, 2014.

10. Pereira J., Xiong Z., et al., Deep Learning for Brain MRI Segmentation: State of the
Art and Future Directions, Journal of Digital Imaging, 2016.

11. Hou J., Samaras D., Multimodal Neuroimaging Feature Learning with Multimodal
Stacked Deep Polynomial Networks for Diagnosis of Alzheimer's Disease, IEEE
Journal of Biomedical and Health Informatics, 2016.

12. Liu M., Zhang D., et al., Multi-modal Discriminative Region Discovery for
Alzheimer's Disease Diagnosis, Neuroinformatics, 2016.

13. Anandhavelu S., Srinivasan K., Convolutional Neural Networks for MRI Brain
Image Classification using Hybrid Feature Selection and Data Augmentation
Techniques, Journal of Medical Systems, 2019.

14. Costa P., Garg A., et al., End-to-End Adversarial Attention Networks for Brain
Tumor Segmentation, Proceedings of the IEEE Conference on Computer Vision
and Pattern Recognition, 2017.

15. Liu M., Zhang D., et al., Multimodal Discriminative Region Discovery for
Alzheimer's Disease Diagnosis, IEEE Journal of Biomedical and Health
Informatics, 2018.

16. Amoroso N., Guaragnella C., et al., Multimodal brain tumor segmentation using
deep convolutional neural networks, International Journal of Computer Assisted
Radiology and Surgery, 2018.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 51

17. Iyappan A., Hojjati A., et al., 3D Convolutional Neural Networks for Classification
of Functional Connectomes in Alzheimer's Disease, Frontiers in Neuroinformatics,
2019.

18. Sarraf S., Tofighi G., et al., Ensemble Deep Learning for Alzheimer's Disease
Diagnosis Using MRI and fMRI Data, Proceedings of the IEEE Conference on
Computer Vision and Pattern Recognition Workshops, 2016.

19. Brosch T., Tang L., et al., Deep 3D Convolutional Encoder Networks with
Shortcuts for Multiscale Feature Integration Applied to Multiple Sclerosis Lesion
Segmentation, Medical Image Analysis, 2016.

20. Suk H., Shen D., et al., Deep Sparse Multi-Modal Learning for Alzheimer's Disease
Diagnosis, International Conference on Medical Image Computing and Computer-
Assisted Intervention, 2013.

21. Anandhavelu S., Srinivasan K., Alzheimer's Disease Detection using Hybrid Deep
Convolutional Neural Networks with MRI Images, Journal of Medical Systems,
2020.

22. Ghafoorian M., Karssemeijer N., et al., Deep Multi-scale Location-aware 3D

Convolutional Neural Networks for Automated Detection of Lacunes of Presumed
Vascular Origin, NeuroImage, 2017.

23. Jafari M., Anssari M., et al., Alzheimer's Disease Detection using Convolutional
Neural Networks and Magnetic Resonance Imaging, Journal of Medical Signals
and Sensors, 2020.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 52

24. Zhao T., Shang W., et al., Discriminative Analysis of Nonlinear Brain Connectivity
for Alzheimer's Disease Identification Using BrainNetCNN, Frontiers in
Neuroscience, 2019.

25. Bernal J., Tajerian M., et al., Multi-view Convolutional Neural Networks for 3D
Shape Classification, IEEE Transactions on Pattern Analysis and Machine
Intelligence, 2017.

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 53

 CHAPTER 10

INFORMATION OF PROJECT GROUP

MEMBERS

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 54

1st Member

• Name - Divya Patil

• Date of Birth - 05 December 2000

• Gender - Female

• Permanent Address - At Post Haidarpur, Tal. Nepanagar, Dist. Burhanpur

• Mobile/Contact No - 7879338580

• Email - divyapatil336@gmail.com

• Placement Details - Atos Global.

• Paper Published- Yes

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 55

2nd Member

• Name- Aishwarya Arun Bharti

• Date of Birth - 27 August 2001

• Gender - Female

• Permanent Address – At Post Nashik, Tal. Nashik, Dist. Nashik

• Mobile/Contact No - 8080423900

• Email – aishbharti145@gmail.com

• Placement Details - No

• Paper Published - Yes

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 56

3rd Member

• Name - Vaishnavi Satish Shelke

• Date of Birth - 20 March 2002

• Gender - Female

• Permanent Address - Vadhane Vasti, Tal. Shrirampur, Dist. Ahmednagar

• Mobile/Contact No - 9307742626

• Email - vaishnavishelke7777@gmail.com

• Placement Details - No

• Paper Published - Yes

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 57

4th Member

• Name - Hemangi Appa Akhade

• Date of Birth - 31 October 1999

• Gender - Female

• Permanent Address - 47, Ganesh Colony, Tal. Shindkheda, Dist. Dhule

• Mobile/Contact No - 9307083801

• Email - hemangiakhade3113@gmail.com

• Placement Details - No

• Paper Published - Yes

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 58

 CHAPTER 11

RESEARCH PAPER

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 59

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 60
SF’s SITRC, Department of Computer Engineering, 2022-23 Page 61
 CHAPTER 12
PAPER PUBLICATION CERTIFICATE

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 62

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 63
SF’s SITRC, Department of Computer Engineering, 2022-23 Page 64
SF’s SITRC, Department of Computer Engineering, 2022-23 Page 65
SF’s SITRC, Department of Computer Engineering, 2022-23 Page 66
SF’s SITRC, Department of Computer Engineering, 2022-23 Page 67
 CHAPTER 13
SPONSORSHIP LETTER

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 68

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 69
SF’s SITRC, Department of Computer Engineering, 2022-23 Page 70
 CHAPTER 14
PLAGIARISM REPORT

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 71

SF’s SITRC, Department of Computer Engineering, 2022-23 Page 72