A team of lecturers and students from Universitas Padjadjaran, in collaboration with several partner universities in Indonesia, has successfully developed LukaKu, a mobile application powered by artificial intelligence (AI) designed to detect external wounds from image data. This research has been published in the international journal Intelligence-Based Medicine.
The LukaKu app is designed to help the public recognize common types of external wounds while also providing first-aid guidance and appropriate medication recommendations. The technology behind the app is YOLOv5, a leading AI model for image-based object detection.
The team tested five YOLOv5 variants (n, s, m, l, and x), and found that YOLOv5l delivered the best performance with a Mean Average Precision (mAP) score of 0.785. This model was then implemented into the mobile app, enabling users to easily access quick and accurate information on wound management.
This study represents a cross-disciplinary collaboration involving the Departments of Electrical Engineering, Mathematics, Statistics, Dermatology and Venereology, and Chemistry at Universitas Padjadjaran, supported by Universitas YARSI, Universitas Jambi, and Sekolah Tinggi Teknologi Bandung.
Beyond offering a practical health solution, this innovation also aligns with the Sustainable Development Goals (SDGs), particularly:
- SDG 3 (Good Health and Well-Being): improving access to healthcare through technology.
- SDG 9 (Industry, Innovation, and Infrastructure): leveraging artificial intelligence for health innovation.
With LukaKu, the public is expected to gain faster, easier, and more reliable access to medical information, especially for initial wound care before seeking treatment at healthcare facilities.
Source: https://www.scopus.com/pages/publications/85215105599
16/Stat/2025
