Deep Learning for Predicting Enhancing Lesions in Mu pie Sclerosis from Noncontrast MRI

Background: Enhancing lesions on MRI scans obtained after contrast material administration are commonly thought to represent disease activity in multiple sclerosis (MS); it is desirable to develop methods that can predict enhancing lesions without the use of contrast material. Purpose: To evaluate whether deep learning can predict enhancing lesions on MRI scans obtained without the use of contrast material. Materials and Methods: This study involved prospective analysis of existing MRI data. A convolutional neural network was used for classification of enhancing lesions on unenhanced MRI scans. This classification was performed for each slice, and the slice scores were combined by using a fully connected network to produce participant-wise predictions. The network input consisted of 1970 multiparametric MRI scans from 1008 patients recruited from 2005 to 2009. Enhanced lesions on postcontrast T1-weighted images served as the ground truth. The network performance was assessed by using fivefold cross-validation. Statistical analysis of the network performance included calculation of lesion detection rates and areas under the receiver operating characteristic curve (AUCs). Results: MRI scans from 1008 participants (mean age, 37.7 years +/- 9.7; 730 women) were analyzed. At least one enhancing lesion was observed in 519 participants. The sensitivity and specificity averaged across the five test sets were 78% +/- 4.3 and 73% +/- 2.7, respectively, for slice-wise prediction. The corresponding participant-wise values were 72% +/- 9.0 and 70% +/- 6.3. The diagnostic performances (AUCs) were 0.82 +/- 0.02 and 0.75 +/- 0.03 for slice-wise and participant-wise enhancement prediction, respectively. Conclusion: Deep learning used with conventional MRI identified enhanced lesions in multiple sclerosis from images from unenhanced multiparametric MRI with moderate to high accuracy. (C) RSNA, 2019