tcama/DL-MS-Gd-Enhancing-Segmentation

A deep learning model foe segmentation of gadolinium-enhancing lesions in multiple sclerosis

3D U-net models for segmentation of gadolinium-enhancing lesions in multiple sclerosis.

Code (using Kera and Tensorflow) and trained model for segmentation of Gd-enhancing lesion from multimodal MRI data of multiple scelrosis patients.

Three models are trained to segment Gd-enhancing lesion:

U1 model: uses post-contrast T1w image

U2 model: uses both pre- and post-contrast T1w images

U2 model: uses PDw, T2w, FLAIR, and pre- and post-contrast T1w images

Please cite the following work:

Deep learning segmentation of gadolinium enhancing lesions in multiple sclerosis

Ivan Coronado, Refaat E. Gabr, Ponnada A. Narayana

Multiple Sclerosis Journal, in press