Jakarta, INTI – Tuberculosis (TB) remains a major challenge for Indonesia’s healthcare system, with estimated deaths reaching 125,000 in 2024. To meet the need for faster and more effective early detection, Gadjah Mada University (UGM) has introduced a technological innovation by developing Indonesia’s first AI-based TB screening application called TBScreen.AI, accessible via http://tbscreen.ai.
Technological Innovation Supports TB Control
According to Wahyono, S.Kom., Ph.D., from UGM’s Department of Computer Science and Electronics, this application is part of the government’s efforts to utilize research and technology results, especially computer-aided diagnosis (CAD) based on AI, to accelerate the TB screening process. The World Health Organization (WHO) has also recommended using CAD as a tool to assist in reading chest X-ray results.
Easy to Use for Medical Personnel and the Public
TBScreen.AI is designed for use by both medical staff and the general public with a simple process: upload a chest X-ray photo that will be automatically analyzed by the AI system. The result is displayed as a percentage likelihood of TB indication, which must then be followed up by a doctor for a definitive diagnosis.
Multidisciplinary Research Collaboration and International Funding
This application development is part of a research project funded by the KONEKSI program from Australia’s Department of Foreign Affairs and Trade (DFAT). The project is led by dr. Antonia Morita I. Saktiawati, Ph.D., Faculty of Medicine, UGM, with Wahyono as AI Team Coordinator. Collaboration also involves institutions such as the University of Melbourne, Monash University Indonesia, and YAKKUM Rehabilitation Center.
Development Process and Data Validation
Initial data came from Sardjito Hospital Yogyakarta and underwent thorough validation by clinical and radiology teams. Using digital image processing and machine learning techniques, the AI model was trained on 936 X-ray data sets, achieving around 64% validity. The team is currently awaiting additional data from Mimika Regional Hospital to improve accuracy.
Key Features and Limited Release
The main feature of the application is automatic screening with an output showing the probability of TB indication between 0–100%. Full access is provided to healthcare workers, while the public can use limited features. The application has been deployed on a limited basis at Klaten Community Health Center and Mimika Regional Hospital as pilot projects to gather feedback before an official release by the end of 2025.
Contribution to Early Detection and Sustainable Development
The presence of TBScreen.AI is expected to accelerate active TB case detection, especially in remote areas and healthcare facilities lacking specialist doctors. This innovation aligns with the government’s TB eradication strategy and supports the achievement of the Sustainable Development Goals (SDGs).
Conclusion:
UGM has successfully developed Indonesia’s first AI-based TB screening application that facilitates early detection of tuberculosis. Through collaborative research and cutting-edge technology, the TBScreen.AI app is expected to accelerate the screening process, support TB eradication efforts, and improve healthcare access, especially in remote areas. This innovation also aligns with national strategies and sustainable development goals.
Read More:Telkomsel Presents 6 Latest AI Solutions at KSTI 2025, Strengthening the Path Toward Golden Indonesia 2045