Rezy Septianda, Kahlil Muchtar, Sayed Muchallil, Maya Fitria, Al Bahri, Novi Maulina
Medication management poses a significant risk for visually impaired (VI) individuals, often leading to medication errors and a dependency on caregivers. To address this, technology-based solutions are crucial for enhancing safety and user autonomy. The proposed approach is a real-time medication identification system developed as a mobile application using the Flutter framework. At its core, the application utilizes a custom-trained You Only Look Once (YOLO) v11 object identification model, which runs entirely on-device to ensure privacy and functionality without an internet connection. The system identifies pharmaceutical packaging from a smartphone's live camera feed and delivers multimodal feedback through a clear voice assistant and haptic vibrations. The model was trained on a custom dataset of 800 images across six medication classes. Evaluation on the validation set demonstrated excellent performance, achieving a mAP@0.5 of 0.978 and a mAP@0.5:0.95 of 0.895. ©2025 IEEE.
Department of Electrical and Computer Engineering, Universitas Syiah Kuala, Aceh, Indonesia; Faculty of Medicine, Universitas Syiah Kuala, Aceh, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock