Utilizing alexnet CNN for enhanced diagnostic accuracy in conjunctivitis detection

Conjunctivitis, commonly known as pink eye, is an inflammation of the thin, transparent layer covering the white part of the eye and the inner surface of the eyelids. It can result from viruses, bacteria, allergens, or irritants, causing symptoms like redness, itching, and discharge. We've identified a challenge in eye health: other diseases can mimic conjunctivitis symptoms such as dry eye syndrome, corneal abrasion, blepharitis and uveitis leading to misdiagnosis and delays in treatment. This highlights the need for accurate diagnostic tools to differentiate between conjunctivitis and similar conditions. Solution: Our approach involves using artificial intelligence, specifically the AlexNet CNN, to detect conjunctivitis from image data. We collected 1032 images, including conjunctivitis and healthy eyes, to train and evaluate the AI model. Employing the AlexNet CNN architecture, we trained the model on the dataset, focusing on optimizing accuracy and performance. Our efforts yielded a remarkable 98% accuracy rate in distinguishing conjunctivitis from healthy eyes, demonstrating the effectiveness of our AI-powered solution. Future plans include expanding the AI system's capabilities to detect a broader range of eye diseases, enhancing its utility in clinical settings and potentially revolutionizing eye healthcare. This innovation could lead to earlier diagnosis, more targeted treatments, and significantly improved patient outcomes.