You are here

• Home
• Best Paper Award at BIOSTEC 2025

Best Paper Award at BIOSTEC 2025

The work of Ouassim Boukhennoufa and his team combines AI and image optimization in nuclear medicine for more accurate and earlier detection of parathyroid anomalies.

Ouassim Boukhennoufa, Jean-Marc Nicod and Noureddine Zerhouni (SUPMICROTECH/FEMTO-ST), in collaboration with Laurent Comas (IH) and Hatem Boulahdour (PUPH) head of the nuclear medicine department at Besançon University Hospital, also members of the SINERGIES laboratory, won the prize for the 2nd best research paper at the 18th Joint International Conference on Systems and Technologies in Biomedical Engineering.

Held in Portugal, in the heart of Porto, this edition of BIOSTEC brought together researchers and practitioners in the fields of engineering, biology, healthcare and informatics. It provided an opportunity to highlight theoretical and applied advances in technologies such as artificial intelligence, signal processing and imaging for biology and medicine.

The award-winning paper proposes a methodology based on AI and medical expertise for the detection of parathyroid adenomas. The aim of this work is to show how improving the quality of nuclear medicine images prior to analysis can lead to better detection of abnormal parathyroid glands. Given the low resolution and high noise of these images, an innovative approach combining image enhancement and AI has been proposed. This method significantly improves the accuracy of classification models, underlining their importance as an aid to medical diagnosis.

This distinction underlines the expertise and innovative contributions of the close collaboration between FEMTO-ST and the Nuclear Medicine Department of Besançon University Hospital in the field of AI for medical imaging and medicine. The award-winning article is a continuation of work already published on this subject.

The research presented in the article is part of Mr. Boukhennoufa's thesis work, which focuses on the integration of artificial intelligence and deterministic approaches to help diagnose hyperparathyroidism, with the aim of guaranteeing better patient care through effective use of the department's technical and analytical resources.

Contact : Ouassim Boukhennoufa et Jean-Marc Nicod