Diagnosis of Alzheimer’s Disease Using Machine Learning Test phase, srs, design phase and source code final deliverable
Domain:
Image processing
Abstract
Alzheimer’s disease is chronic condition that leads to degeneration of brain cells leading at memory loss. Patients with cognitive mental problems such as confusion and forgetfulness, also other symptoms including behavioral and psychological problems are further suggested having CT, MRI, PET, EEG, and other neuroimaging techniques.
The aim of this paper is making use of machine learning algorithms to process this data obtained by neuroimaging technologies for detection of Alzheimer’s in its primitive stage.
Functional Requirements:
The workflow of the project is shown in the diagram below:
- Image data acquisition and perform preprocessing of the data.
- Use Brouta Algorithm (available with python package) for feature selection.
- Apply any suitable machine learning (supervised learning) algorithm for classification.
- The classification results were calculated by means of three metric measurements, which are used for quantitative valuation and evaluation, including accuracy, sensitivity (recall) and specificity. Additionally, numerous optimal approaches such as receiver operating curve (ROC) and Area under the Curve (AUC) are calculated as well.
- Use at least two algorithms like support vector machine and Linear Regression and do comparison to select the best one the basis of performance measures.
Helping Material:
You can consult the following links for better understanding of the project: For dataset:
https://www.oasis-brains.org/#data
Research articles related to the projects:
- https://iopscience.iop.org/article/10.1088/1742-6596/1921/1/012024
- https://ieeexplore.ieee.org/document/8697386
- https://iopscience.iop.org/article/10.1088/1742-6596/1372/1/012065/pdf
Tools:
You can use any of the following tools Matlab, Weka, Python
Supervisor:
Name: Noureen Hameed
