Development of an AI Tool for Early Detection of Liver Cancer in Benin

Vincent-Béni Sèna Zossou, a researcher at CESP of the University of Versailles Saint-Quentin-en-Yvelines, has developed an artificial intelligence tool capable of analyzing CT scan images to detect liver cancer. This innovation, validated in 2024, promises to improve diagnostics in West Africa, where hepatocellular carcinoma is often diagnosed late.

Early detection of hepatocellular carcinoma (HCC) is a public health issue in West Africa, where late diagnoses lead to high mortality rates. Traditional methods, based on visual analysis of CT scan images, are often prone to errors. Vincent-Béni Sèna Zossou offers an innovative solution by integrating artificial intelligence to automate this process.

The tool developed by Zossou relies on the analysis of 1,764 studies and 77 articles to optimize its algorithms. The results show Dice scores for the segmentation of liver lesions ranging from 69% to 97.1%, demonstrating the effectiveness of AI in the early diagnosis of HCC, thus reducing the time needed to analyze images.

This advancement represents a strategic opportunity for public health decision-makers. By improving diagnostic tools, it is possible to increase early detection rates of liver cancer, which could transform medical practices in Africa. Investors are encouraged to support the development of these technologies to meet local needs.

« « Thanks to our AI tool, we have achieved Dice scores of up to 97.1% for the segmentation of liver lesions, which could transform the diagnosis of liver cancer in Africa. » »

« « AI innovation for medical diagnosis is essential to improve access to quality care in Africa, and we must support these initiatives for a sustainable impact. » »