Africa (Commonwealth Union) _ At the Kenya Medical Research Institute (KEMRI), significant progress is being made in the development of a mobile phone application that utilizes artificial intelligence (AI) to diagnose tuberculosis (TB) and other respiratory diseases. Led by Dr. Videlis Nduba and his team, the research aims to differentiate between coughs associated with respiratory diseases like TB and those from individuals without such conditions.
In a controlled environment, cough samples are collected from individuals with respiratory diseases and those without, using various microphones, including ones integrated into smartphones. These recordings are then analyzed using existing computer software called ResNet 18, which employs AI to distinguish between different cough patterns.
Dr. Nduba highlights the potential impact of the software in reducing the time taken for diagnosis and treatment of TB. He emphasizes that early diagnosis can significantly curb the spread of the disease, as individuals can be promptly isolated and treated, thus reducing transmission within communities.
However, the software’s current accuracy levels do not meet the standards set by the World Health Organization (WHO). While the WHO requires a minimum accuracy of 90% in detecting TB infections and 80% accuracy in ruling out the presence of TB, the trials conducted so far have demonstrated 80% accuracy for TB detection and 70% accuracy for ruling out TB.
Despite these challenges, participants like former TB patient Johnson Munori are optimistic about the potential of the technology. Munori highlights the convenience and effectiveness of the app in his own diagnosis and treatment journey.
Jarim Omogi, a specialist in public health at Amref International University, underscores the growing role of AI in medicine and its potential to expedite patient care. He emphasizes the importance of focusing not just on diagnosis but also on factors such as treatment efficiency, cost-effectiveness, and real-time data accessibility.
The trial, funded by the National Institutes of Health (NIH), represents a significant step forward in leveraging technology to address public health challenges. However, regulatory approval is still pending, underscoring the need for further validation and refinement of the technology before widespread implementation.
https://www.who.int/news-room/fact-sheets/detail/tuberculosis
