For full functionality of this site it is necessary to enable JavaScript. Here are the instructions how to enable JavaScript in your web browser.

Scientific Program

FP12

1405

Dr.Mahesh Babu
Membership No Author Name Email Mobile
4441MW Dr.Priyanka Kanekal priyankakanekal84@gmail.com .
Medical Retina

Evaluation  of Artificial  Intelligence in Image  Based Diagnosis of Posterior  Segment  Pathologies

Purpose: To evaluate the diagnostic performance of artificial intelligence (AI) in detecting posterior segment pathologies from retinal images and compare its accuracy with ophthalmologist assessment.

Methods: A cross-sectional study was conducted on 10 retinal fundus images with various posterior segment disorders. AI-generated diagnoses were compared with expert ophthalmologist diagnoses,  Diagnostic accuracy, sensitivity, specificity, and agreement rates were analyzed.

Results: ChatGPT demonstrated a diagnostic agreement of 73.3%, compared with 53.3% for Gemini. Mean diagnostic accuracy was 82.4% for ChatGPT and 64.8% for Gemini, while ophthalmologists achieved 98.6% accuracy. 

Conclusion: Despite limited concordance, AI models failed to achieve the diagnostic reliability of ophthalmologists and should not be used for clinical diagnosis or decision making.

Membership No Author Name Email Mobile
379 Prof.Sudhakar N A sudhakarappaji11@yahoo.co.in 9243393178