This community innovation has been accepted at the 2025 DHIS2 Annual Conference
AI-Based Detection of Tuberculosis in HIV Patients
Abstract HIV and Tuberculosis (TB) co-infection remains a significant public health challenge in Pakistan, particularly in resource-limited settings. Timely and accurate detection is critical for improving patient outcomes and reducing the disease burden. This intervention leverages DHIS2 for managing national health data and integrates machine learning techniques to enhance the detection of TB co-infection in HIV patients. By utilizing structured datasets containing patient demographics, clinical history, and lab results, machine learning models improve diagnostic accuracy and provide actionable insights. Our approach streamlines data harmonization, visualization, and analysis through DHIS2’s flexible and real-time reporting tools. This system enables public health professionals to monitor disease trends efficiently, make evidence-based decisions, and optimize resource allocation for HIV, TB, and Malaria programs. Additionally, the integration of machine learning with DHIS2 supports predictive analytics, facilitating early detection and proactive interventions. The impact of this intervention includes improved programmatic outcomes, reduced disease burden, enhanced collaboration between stakeholders, and strengthened health systems. This approach serves as a scalable and replicable model for other countries and organizations tackling similar health challenges. By sharing our experiences, we aim to inspire global efforts to adopt advanced technologies like DHIS2 and machine learning to improve healthcare delivery, foster collaboration, and achieve better public health outcomes.
Primary Author: Muhammad Babar
Keywords:
Tuberculosis (TB), Human Immunodeficiency Virus (HIV), Machine Learning (ML), Disease Detection, Healthcare Analytics, Public Health