Published Research Paper

This research explores how convolutional neural networks (CNNs) can improve pill image recognition for medical retrieval systems, helping patients and practitioners identify unidentified medications through camera-equipped mobile devices.

The study proposes three CNN architectures: two hybrid models (CNN+SVM and CNN+kNN) and a ResNet-50 network. Various preprocessing steps using detection techniques including Gaussian filtering were applied to a dataset of 7,000 pill images from the National Library of Medicine.

The proposed CNN+kNN architecture achieved 90.8% accuracy, a roughly 10% improvement over existing methods, with a runtime of approximately 1 millisecond per execution.

I co-authored this paper with faculty at Rochester Institute of Technology, Dubai. My contributions included designing the methodology, running model training and evaluations, and documenting the experimental results.

This work deepened my understanding of applied machine learning and strengthened my ability to bridge computer science with real-world problems in healthcare technology.