Data Challenge_TB-FUNGI DETECT-AI SYSTEM

Background: Antimicrobial resistance (AMR) is a critical global health issue, severely affecting healthcare provision. In Uganda, AMR impacts more than 74% of patients, resulting in over 1.27 million deaths annually. Contributing factors include self-medication, medical errors, and inappropriate antimicrobial use. Similarly, tuberculosis (TB) and various fungal infections have shown increased resistance to conventional treatments, complicating their management. Medication errors in African settings account for 8.4% of patient deaths, and over 8 million individuals in low- and middle-income countries (LMICs) die each year from treatable AMR conditions. These fatalities contradict Sustainable Development Goal 3, necessitating innovative approaches to healthcare systems and diagnostics to reduce mortality rates. Summary of Proposed Research: To address this urgent need, we propose leveraging machine learning (ML) and artificial intelligence (AI) for the diagnosis of tuberculosis and fungal infections. ML enables computer systems to learn and adapt without explicit instructions, while AI replicates human-like intelligence. AI-assisted approaches have shown promise in improving diagnostic accuracy, reaching rates as high as 96.7%. Our research focuses on developing an AI system, named TB-Fungi DetectAI, for quick and accurate self-diagnosis, particularly in resource-limited LMICs. Using a decision tree-based ML approach, we shall create a diagnostic system by leveraging repositories, developing databases, and designing user interfaces. The Clinical Decision Support System (CDSS) will assist doctors in diagnosing TB and fungal infections by comparing subtle symptoms against a knowledge base, offering non-invasive diagnosis only when a match is found. This reduces false positives and negatives associated with traditional methods. Future directions involve expanding the AI system into a standalone platform to expedite diagnosis and address the scarcity of medical specialists in LMICs, including Uganda. This research will improve patient outcomes, strengthen stewardship and health systems, and inform public health strategies by showcasing the potential of AI-assisted diagnostics in combating TB, fungal infections, and AMR