Meeting Banner
Abstract #1399

Robust multi-atlas MRI segmentation with corrective learning for quantification of local quadriceps muscles inflammation changes during a longitudinal study in athletes

Hoai-Thu Nguyen1, Pierre Croisille1,2, Magalie Viallon 1,2, Charles de Bourguignon2, Rémi Grange2, Sylvain Grange1,2, and Thomas Grenier3

1Univ Lyon, UJM-Saint-Etienne, INSA-Lyon, Université Claude Bernard Lyon 1, CNRS, Inserm, CREATIS UMR 5220, U1206, F-42023, Saint-Etienne, France, 2Department of Radiology, Centre Hospitalier Universitaire de Saint-Etienne, Université Jean-Monnet, Saint-Etienne, France, 3Univ Lyon, INSA‐Lyon, Université Claude Bernard Lyon 1, UJM-Saint Etienne, CNRS, Inserm, CREATIS UMR 5220, U1206, F-69621, Villeurbanne, France

This study propose an improved automatic segmentation of longitudinal MRI dataset of mountain ultra-marathon runners’ upper thighs acquired during the Tor des Géants 2014 by using a multi-atlas segmentation strategy with corrective learning with a small number of training set. Our highly accurate and robust segmentations allow us to locally study the inflammation of each quadriceps head induced by the extreme conditions of the race, a method that is of high interest to monitor the impact of eccentric efforts during the race, identify local physiopathology changes in patients, and benefits of eventual therapy or intervention.

This abstract and the presentation materials are available to members only; a login is required.

Join Here