ALL-Net: Anatomical information lesion-wise loss function integrated into neural network for multiple sclerosis lesion segmentation

•A new algorithm, ALL-Net, is introduced to improve MS lesion segmentation.•ALL-Net integrates anatomical coordinate information into the neural network.•ALL-Net integrates lesion-wise loss function to improve small lesion detection.•ALL-Net is robust to both small- and large-scale MS lesion dataset...

Full description

Saved in:
Bibliographic Details
Published in:NeuroImage clinical Vol. 32; p. 102854
Main Authors: Zhang, Hang, Zhang, Jinwei, Li, Chao, Sweeney, Elizabeth M., Spincemaille, Pascal, Nguyen, Thanh D., Gauthier, Susan A., Wang, Yi, Marcille, Melanie
Format: Journal Article
Language:English
Published: Netherlands Elsevier Inc 01.01.2021
Elsevier
Subjects:
Algorithms
Convolutional neural network
Deep learning
Humans
Image Processing, Computer-Assisted
Lesion segmentation
Magnetic Resonance Imaging
Mental Recall
MRI
Multiple sclerosis
Multiple Sclerosis - diagnostic imaging
Neural Networks, Computer
Radiology
Regular
Deep learning
Multiple sclerosis
Lesion segmentation
MRI
Convolutional neural network
ISSN:2213-1582, 2213-1582
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A new algorithm, ALL-Net, is introduced to improve MS lesion segmentation.•ALL-Net integrates anatomical coordinate information into the neural network.•ALL-Net integrates lesion-wise loss function to improve small lesion detection.•ALL-Net is robust to both small- and large-scale MS lesion datasets. Accurate detection and segmentation of multiple sclerosis (MS) brain lesions on magnetic resonance images are important for disease diagnosis and treatment. This is a challenging task as lesions vary greatly in size, shape, location, and image contrast. The objective of our study was to develop an algorithm based on deep convolutional neural network integrated with anatomic information and lesion-wise loss function (ALL-Net) for fast and accurate automated segmentation of MS lesions. Distance transformation mapping was used to construct a convolutional module that encoded lesion-specific anatomical information. To overcome the lesion size imbalance during network training and improve the detection of small lesions, a lesion-wise loss function was developed in which individual lesions were modeled as spheres of equal size. On the ISBI-2015 longitudinal MS lesion segmentation challenge dataset (19 subjects in total), ALL-Net achieved an overall score of 93.32 and was amongst the top performing methods. On the larger Cornell MS dataset (176 subjects in total), ALL-Net significantly improved both voxel-wise metrics (Dice improvement of 3.9% to 35.3% with p-values ranging from p < 0.01 to p < 0.0001, and AUC of voxel-wise precision-recall curve improvement of 2.1% to 29.8%) and lesion-wise metrics (lesion-wise F1 score improvement of 12.6% to 29.8% with all p-values p < 0.0001, and AUC of lesion-wise ROC curve improvement of 1.4% to 20.0%) compared to leading publicly available MS lesion segmentation tools.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:2213-1582
2213-1582
DOI:10.1016/j.nicl.2021.102854