Table 1

- Features of included studies

ArticleCountryDatasetSample sizeDiagnosisApplicationDeep learning architecturePerformance
Afzal el al, 201815AustraliaJohn Hunter Hospital’s Dataset21-11 converted to MS-10 did not convert to MSClassification2D-CNNAccuracy
Afzal el al, 202116AustraliaISBI and MICCAI datasets19127 scans of MSSegmentation2D-CNN-DSC-Sensitivity-Precision
Alijamaat et al, 202017IranLaboratory of eHealth of the University of Cyprus5838 MS patients 20 healthy individualsIdentification2D-CNN-Accuracy-Precision-Sensitivity-Specificity
Aslani et al., 201918Italy-Private dataset-ISBI 2015 longitudinal dataset51-37 patients from private dataset -14 patients from ISBI 2015 longitudinal datasetSegmentation2D-CNNDSC
Aslani et al, 201919ItalyISBI 2015 Longitudinal MS Lesion Segmentation19MSSegmentation2D-CNN-DSC-Lesion-wise true-positive -Lesion-wise false-positive
Coronado et al, 202020USACombiRx1,006Relapsing–remitting MSSegmentation3D-CNN-DSC-Lesion-wise true-positive-Lesion-wise false-positive
Eitel et al, 201921GermanyClinical14776 MS patients 71 healthy patientsClassification3D-CNNAccuracy
Kazancli et al, 201822SpainClinical59MSSegmentation3D-CNN-DSC-True Positive Rate-False Discovery Rate-Volume Difference
La Rosa et al, 201823SwitzerlandClinical105-Training dataset: 32 patients with EDSS scores ranged from 1 to 2
-Test dataset: 73 patients with EDSS scores ranged from 1 to 7.5
Segmentation3D-CNN-DSC-Lesion-wise false positive-Lesion-wise true positive-Volume difference
Roy et al, 201824USAISBI 201519-Training dataset: 5 patients with MS -Test dataset: 14 patients with MSSegmentation2D-CNNDSC
Shrwan et al, 202125IndiaClinical38MSClassification2D-CNN-Accuracy-Precision-Recall f_score
Siar et al, 201926IranClinical1111320 MS patients 791 healthy patientsClassification2D-CNN-Accuracy-Sensitivity-Specificity
Valverde et al, 201827SpainMICCAI 2008 MICCAI 2016 ISBI 201560MSSegmentation3D-CNN-DSC-Sensitivity-Precision
Wang et al, 201828ChinaeHealth Laboratory and Private data6438 MS patients 26 healthy patientsIdentification2D-CNN-Accuracy-Sensitivity-Specificity
Zhang et al, 201829ChinaeHealth Laboratory and Private data6438 MS patients 26 healthy patientsIdentification3D-CNN-Accuracy-Sensitivity-Specificity
  • CNN: convolutional neural network, CombiRx: Combination Therapy in Patients with Relapsing-Remitting Multiple Sclerosis, DSC: Dice Similarity Coefficient, EDSS: Expanded Disability Status Scale, ISBI: International Symposium on Biomedical Imaging, MICCAI: Medical Image Computing and Computer Assisted Intervention, MS: Multiple sclerosis.