Rédaction Africa Links 24 with Africanews
Published on 2024-02-25 16:49:04
The Kenya Medical Research Institute is currently conducting research to develop a mobile phone application that utilizes artificial intelligence (AI) to diagnose tuberculosis and other respiratory diseases. Dr. Videlis Nduba and his team are working in a specially designed room to record coughs from individuals with respiratory diseases such as tuberculosis, as well as those without any disease. The objective is to create software that can distinguish between the two and ultimately develop a mobile phone application that can accurately identify a cough associated with TB and other serious illnesses.
The team collects natural or forced coughs using three different microphones, including a low-cost version, a high-definition one, and a smartphone microphone. The data collected is then sent to the University of Washington, where it is analyzed using a computer software system called ResNet 18. Dr. Nduba, the principal researcher, explains that the software utilizes artificial intelligence to analyze the coughs and determine if there are differences between individuals with TB and those without the disease.
If the software proves to be accurate in clinical trials, it has the potential to significantly reduce the time taken to diagnose and treat patients with TB, ultimately helping to control the spread of the disease. Dr. Nduba emphasizes that a quicker diagnosis can reduce TB transmission within communities, as patients can receive treatment sooner.
While the software has shown promising results, it has not yet met the accuracy standards set by the World Health Organization (WHO). The WHO requires that any diagnostic application for TB must be at least 90% accurate in detecting the infection and at least 80% accurate in ruling out the presence of the disease. Dr. Nduba’s trials have shown 80% accuracy in detecting TB and 70% accuracy in ruling out the disease, falling slightly short of the WHO’s requirements.
Former TB patient, Johnson Munori, who participated in the research, believes that the mobile application could be beneficial in diagnosing TB more efficiently. He shares his experience of being diagnosed with TB through the research program and how the new technology could help many people receive timely treatment.
Jarim Omogi, a public health specialist at Amref International University, highlights the growing use of AI in medicine and the potential benefits it offers in terms of faster diagnosis and treatment. He emphasizes the importance of considering the speed, cost-effectiveness, and efficiency of such technologies in improving patient outcomes.
The research trial is currently funded by the National Institutes of Health, but it has not yet received regulatory approval. The ongoing work at the Kenya Medical Research Institute showcases the potential of AI-powered applications in revolutionizing the diagnosis and treatment of infectious diseases like tuberculosis.
Read Original article on Africa News
