Deep learning enhances metagenomics for disease diagnosis in Benin

The thesis of Thanh Hai Nguyen, defended at Gaston Berger University in Saint-Louis, reveals how deep learning transforms the analysis of human microbiomes, thereby facilitating the diagnosis and prognosis of diseases. This research, conducted in 2018, proposes an innovative method for extracting biomedical signatures, opening new perspectives for public health in Benin.

Thanh Hai Nguyen's research focuses on the application of deep learning to metagenomics, a discipline that analyzes the entire genomes of an environmental sample. Microbiomes, influenced by various factors, pose analytical challenges. This thesis aims to overcome these obstacles by developing a feature selection framework to extract relevant information from complex data.

Nguyen established a framework using self-organizing maps to visualize and merge diverse data, allowing for accurate classification of microbiomes. Although specific figures are not available, this approach has demonstrated increased effectiveness in identifying potential biomarkers for diseases, highlighting the importance of microbiome analysis in medical diagnosis.

The results of this research have strategic implications for public health decision-makers, providing advanced analytical tools for disease diagnosis. Recommendations include integrating these technologies into clinical practices and promoting partnerships between startups and research institutions to maximize the impact of these innovations on health in Benin.

« « The integration of deep learning in metagenomics allows for the identification of potential biomarkers for specific diseases, which could transform medical diagnosis. » »

« « This research highlights the importance of accessing advanced knowledge to improve public health decisions and technological innovation. » »