Scientists have developed an AI-enabled diagnostic assistant that can help identify suspected leprosy lesions.
The tool, called AI4Leprosy, works by assessing skin lesion images in combination with patient symptoms.
A paper published in the Lancet Regional Health – Americas shows the tech is over 90 per cent accurate in detecting leprosy.
According to the authors, the publication provides proof of concept for the method, which now serves as the basis for developing a mobile application to further validate the tool, and ultimately accelerate leprosy detection globally.
Ann Aerts, head of the Novartis Foundation, said: “The results are an exciting indication of the potential of AI4Leprosy. We are convinced that its further validation and global rollout can help cover the last miles toward leprosy elimination and use the newest technology available to end one of the oldest scourges known to man.
“The near elimination of leprosy is one of the greatest public health successes in history. Free availability of multidrug therapy (MDT) – donated by Novartis to all leprosy patients worldwide through the World Health Organization (WHO) – has been a game changer on the path to elimination, reducing the global burden of leprosy by 95 per cent over the past decades.
“However, delays in diagnosing the disease allows leprosy to spread, and for the past ten years, the number of newly diagnosed patients has stagnated at about 200,000 per year.”
The authors of the study published in Lancet Americas believe AI can contribute to interrupting transmission and to end this historical disease once and for all.
Milton Ozório Moraes, head of the leprosy laboratory at Fiocruz, said: “Currently, leprosy diagnosis takes place in health facilities, often by non-specialised health professionals. Our goal is to produce an intuitive application where health professionals can take a skin lesion image and provide some clinical information, send it to the cloud and receive the probability of leprosy.”
Juan Lavista Ferres, Microsoft’s chief data scientist, said: “Leprosy is one of humanity’s oldest diseases, although it can be cured if detected early. By leveraging AI and machine learning, patients can get to the right experts at the right time, potentially improving the lives of millions of people around the world. The possibilities with this kind of application are endless, especially for multi-faceted diseases such as leprosy.”
Fiocruz, Microsoft’s AI for health team, the Novartis Foundation and the Universities of Basel and Aberdeen, participated in the study.
