Differentiation between multiple sclerosis and neuromyelitis optica spectrum disorders by multiparametric quantitative MRI using convolutional neural network

Abstract

Multiple sclerosis and neuromyelitis optica spectrum disorders are both neuroinflammatory diseases and have overlapping clinical manifestations. We developed a convolutional neural network model that differentiates between the two based on magnetic resonance imaging data. Thirty-five patients with relapsing-remitting multiple sclerosis and eighteen age-, sex-, disease duration-, and Expanded Disease Status Scale-matched patients with anti-aquaporin-4 antibody-positive neuromyelitis optica spectrum disorders were included in this study. All patients were scanned on a 3-T scanner using a multidynamic multi-echo sequence that simultaneously measures R1 and R2 relaxation rates and proton density. R1, R2, and proton density maps were analyzed using our convolutional neural network model. To avoid overfitting on a small dataset, we aimed to separate features of images into those specific to an image and those common to the group, based on SqueezeNet. We used only common features for classification. Leave-one-out cross validation was performed to evaluate the performance of the model. The area under the receiver operating characteristic curve of the developed convolutional neural network model for differentiating between the two disorders was 0.859. The sensitivity to multiple sclerosis and neuromyelitis optica spectrum disorders, and accuracy were 80.0%, 83.3%, and 81.1%, respectively. In conclusion, we developed a convolutional neural network model that differentiates between multiple sclerosis and neuromyelitis optica spectrum disorders, and which is designed to avoid overfitting on small training datasets. Our proposed algorithm may facilitate a differential diagnosis of these diseases in clinical practice.

Multimodal magnetic resonance imaging quantification of gray matter alterations in relapsing‐remitting multiple sclerosis and neuromyelitis optica spectrum disorder

“…This was a retrospective case-control study comprising 66 subjects matched for age and sex (Table 1), including 30 patients (6 males and 24 females, 51.43 ± 8.02 years old) with a diagnosis of RRMS according to the revised McDonald criteria 2017 (Thompson et al, 2018), 18 patients (2 males and 16 females, 52.67 ± 16.07 years old) with a diagnosis of NMOSD according to Wingerchuk's criteria 2015 (Wingerchuk et al, 2015), and 19 healthy controls (6 males and 13 females, 51.47 ± 9.25 years old). The population of this study overlapped with that of a previously published study (Hagiwara et al, 2021). All patients with NMOSD were anti-AQP4 IgG seropositive, whereas all patients with RRMS were seronegative.…”

Multiparametric MRI

“…Convolutional neural network (CNN) is a type of deep learning known to be useful for image tasks, with its architecture resembling that of the human visual cortex 128,129 . In recent years, deep learning, particularly CNN, has gained substantial popularity in the MRI field for image reconstruction, image quality improvement, image transfer, disease detection, tissue segmentation, and classification 130–142 …”

“…128,129 In recent years, deep learning, particularly CNN, has gained substantial popularity in the MRI field for image reconstruction, image quality improvement, image transfer, disease detection, tissue segmentation, and classification. [130][131][132][133][134][135][136][137][138][139][140][141][142] Multiparametric MRI has been used for analyses using deep learning. Multiparametric MRI is potentially useful for lesion segmentation to fully capture the extent of the disease and has been used to segment cancers and multiple sclerosis plaques.…”

“…Multiparametric MRI has also been used for the classification or diagnosis of diseases using deep learning. 135,[150][151][152][153][154][155][156][157][158] Hagiwara et al 135 applied CNN to T1, T2, and proton density maps of the brain to extract features common to multiple sclerosis and neuromyelitis optica spectrum disorder, another demyelinating disease, and used this information to differentiate these 2 disorders, achieving an accuracy of 80%. Some studies showed the superiority of the diagnostic ability of CNN over radiomics based on multiparametric MRI.…”

Research Progress of Artificial Intelligence Image Analysis in Systemic Disease-Related Ophthalmopathy

“…After 10 times cross-validation and independent testing, the results showed that in training, the accuracy of the MM-RF model was 0.849, and the AUC value was 0.826; for testing, the accuracy of the MM-RF model was 0.871, and the AUC value was 0.902. Hagiwara et al [38] constructed a convolution neural network model based on SqueezeNet to distinguish between NMO and MS. In this study, 35 patients with MS and 18 patients with NMO were enrolled, and left-over cross-validation (leave-one-out cross validation) was used to evaluate the performance of the model.…”