fr | en
Laboratoire Angevin de Recherche en Ingénierie des Systèmes

Séparés par des virgules

Séminaires des doctorants - Lukman Enegi Ismaila14h00 | Polytech Angers | B106 | 62 Notre Dame du Lac - 49000 Angers

Sujet : Machine learning application in neuroscience for neurosurgical brain tumor resection procedure.


This study discusses the challenges of identifying functional brain areas during brain tumor resection surgery. Task-based paradigms, such as finger tapping or speech, are traditionally used, but this approach is not suitable for all patients among other limitations. Resting state functional magnetic resonance imaging (rs-fMRI) is an attractive alternative, but it is not widely used in clinical practice due to the need for manual recognition of functional brain networks. In effort to promote the standardization of this technique, we demonstrated brain network activation feature transferability to improve the classification accuracy of unhealthy data through our proposed end-to-end deep learning algorithm for automatic functional brain networks recognition. We also illustrated the use of self-supervision learning to avoid healthy data annotation since it is non-relevant in clinical procedure and lastly, we proposed a more frugal approach for 3D image data modeling to avoid large training parameters in the functional brain network identification task.