Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/601378
Title: | Eye Movement Analysis of Glaucoma Patients using Eye tracker Device |
Researcher: | Sajitha Krishnan |
Guide(s): | Amudha J |
Keywords: | Computer Science Artificial Intelligence Engineering and Technology Visual impairment;Glaucoma diseases; glaucoma screening; eye movement analysis ; eye care; gaze patterns glaucoma; machine learning; neural network model ; rural healthcare ; glaucoma detection algorithms; gaze analysis; fusion maps eye tracking; Artificial Intelligence; ophthalmology; Optometry; Visual Sciences; |
University: | Amrita Vishwa Vidyapeetham University |
Completed Date: | 2024 |
Abstract: | Glaucoma is a group of diseases causing irreversible optic nerve damage, necessitating newlineearly detection through regular eye exams. Primary eye care centers in rural newlineIndia lack glaucoma screening due to cost and technical constraints. Understanding newlineglaucoma progression is challenging in rural areas where diagnostic equipment newlineand expertise are lacking. There s a significant need to deploy a glaucoma assessment system in primary eye care centers. Although researchers have created less costly open-source glaucoma screening software, there exists a challenge in developing newlinea complete system that employs machine learning models to differentiate newlinebetween glaucoma and normal conditions. To address this, a low-cost, portable newlineglaucoma screening tool is proposed, incorporating an eye tracker for assessing eye movement patterns, which can serve as a biomarker for diagnosis. The thesis, titled Eye movement analysis of glaucoma patients using an eye tracker device, aligns with creating a glaucoma assessment system. The focus on eye movement analysis implies a systematic and analytical approach to understand the gaze patterns in glaucoma patients, which contributes to develop an assessment system. The primary contribution of the research is the Eye gaze dataset, which is the key resource for the thesis. The thesis has investigated multiple algorithms newlineto calculate derived parameters from unprocessed eye gaze data, which are the newlinefoundation for subsequent analysis and categorization. newlineThe thesis proposes a system consisting of four distinct frameworks. The primary newlineversion is the Gaze Fusion Neural Network Model (GFDM), which utilizes fusion newlinemaps derived from eye-tracking data to detect glaucoma. The GFDM employs newlinea transfer learning approach with ImageNet weights to extract features, enabling newlineit to differentiate between normal and glaucomatous conditions. Additionally, the newlinesystem incorporates visualizations such as the Gaze Fusion Map (GFRT), Gaze newlineConvex Hull Map (GCHM), and Gaze Fusion Reaction Time (GFRT) map to newlineillustrate ... |
Pagination: | xiv, 159 |
URI: | http://hdl.handle.net/10603/601378 |
Appears in Departments: | Amrita School of Computing |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 337.19 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 879.16 kB | Adobe PDF | View/Open | |
03_certificate of plagiarism_sajitha krishnan.pdf | 301.74 kB | Adobe PDF | View/Open | |
04_contents.pdf | 68.66 kB | Adobe PDF | View/Open | |
05_abstract.pdf | 53.29 kB | Adobe PDF | View/Open | |
06_chapter 1.pdf | 7.63 MB | Adobe PDF | View/Open | |
07_chapter 2.pdf | 3.77 MB | Adobe PDF | View/Open | |
08_chapter 3.pdf | 9.44 MB | Adobe PDF | View/Open | |
09_chapter 4.pdf | 1.86 MB | Adobe PDF | View/Open | |
10_chapter 5.pdf | 1.1 MB | Adobe PDF | View/Open | |
12_chapter 7.pdf | 1.26 MB | Adobe PDF | View/Open | |
14_annexure.pdf | 171.63 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 350.57 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: