QSMRim-Net: Imbalance-aware learning for identification of chronic active multiple sclerosis lesions on quantitative susceptibility maps.

TitleQSMRim-Net: Imbalance-aware learning for identification of chronic active multiple sclerosis lesions on quantitative susceptibility maps.
Publication TypeJournal Article
Year of Publication2022
AuthorsZhang H, Nguyen TD, Zhang J, Marcille M, Spincemaille P, Wang Y, Gauthier SA, Sweeney EM
JournalNeuroimage Clin
Volume34
Pagination102979
Date Published2022
ISSN2213-1582
KeywordsAlgorithms, Brain, Humans, Magnetic Resonance Imaging, Multiple Sclerosis, Neural Networks, Computer
Abstract

BACKGROUND AND PURPOSE: Chronic active multiple sclerosis (MS) lesions are characterized by a paramagnetic rim at the edge of the lesion and are associated with increased disability in patients. Quantitative susceptibility mapping (QSM) is an MRI technique that is sensitive to chronic active lesions, termed rim + lesions on the QSM. We present QSMRim-Net, a data imbalance-aware deep neural network that fuses lesion-level radiomic and convolutional image features for automated identification of rim + lesions on QSM.

METHODS: QSM and T2-weighted-Fluid-Attenuated Inversion Recovery (T2-FLAIR) MRI of the brain were collected at 3 T for 172 MS patients. Rim + lesions were manually annotated by two human experts, followed by consensus from a third expert, for a total of 177 rim + and 3986 rim negative (rim-) lesions. Our automated rim + detection algorithm, QSMRim-Net, consists of a two-branch feature extraction network and a synthetic minority oversampling network to classify rim + lesions. The first network branch is for image feature extraction from the QSM and T2-FLAIR, and the second network branch is a fully connected network for QSM lesion-level radiomic feature extraction. The oversampling network is designed to increase classification performance with imbalanced data.

RESULTS: On a lesion-level, in a five-fold cross validation framework, the proposed QSMRim-Net detected rim + lesions with a partial area under the receiver operating characteristic curve (pROC AUC) of 0.760, where clinically relevant false positive rates of less than 0.1 were considered. The method attained an area under the precision recall curve (PR AUC) of 0.704. QSMRim-Net out-performed other state-of-the-art methods applied to the QSM on both pROC AUC and PR AUC. On a subject-level, comparing the predicted rim + lesion count and the human expert annotated count, QSMRim-Net achieved the lowest mean square error of 0.98 and the highest correlation of 0.89 (95% CI: 0.86, 0.92).

CONCLUSION: This study develops a novel automated deep neural network for rim + MS lesion identification using T2-FLAIR and QSM images.

DOI10.1016/j.nicl.2022.102979
Alternate JournalNeuroimage Clin
PubMed ID35247730
PubMed Central IDPMC8892132
Grant ListR01 NS104283 / NS / NINDS NIH HHS / United States
UL1 TR002384 / TR / NCATS NIH HHS / United States

Weill Cornell Medicine Neurology 525 E. 68th St.
PO Box 117
New York, NY 10065 Phone: (212) 746-6575