Comparing lesion segmentation methods in multiple sclerosis: Input from one manually delineated subject is sufficient for accurate lesion segmentation

• Published in: NeuroImage clinical 2019-01, Vol.24, p.102074-102074, Article 102074
• Main Authors: Weeda, M.M., Brouwer, I., de Vos, M.L., de Vries, M.S., Barkhof, F., Pouwels, P.J.W., Vrenken, H.
• Format: Article
• Language: English
• Subjects: Adult; Automatic lesion segmentation; Brain - diagnostic imaging; Convolutional neural networks; Deep Learning; Female; Humans; Image Processing, Computer-Assisted; Magnetic Resonance Imaging; Male; Middle Aged; MRI; Multiple sclerosis; Multiple Sclerosis, Relapsing-Remitting - diagnostic imaging; Neural Networks, Computer; Regular; Supervised Machine Learning; Unsupervised Machine Learning
• ISSN: 2213-1582; 2213-1582
• DOI: 10.1016/j.nicl.2019.102074

•Lesion segmentation is an important tool in multiple sclerosis (MS) research.•Several automated methods exist, e.g., supervised, trained or deep learning-based.•Convolutional neural networks (CNN) show promising results for lesion segmentation.•CNN method nicMSlesions can be trained with input from one single-subject.•With minimal input, nicMSlesions outperforms e.g. LST-LPA and LesionTOADS. Accurate lesion segmentation is important for measurements of lesion load and atrophy in subjects with multiple sclerosis (MS). International MS lesion challenges show a preference of convolutional neural networks (CNN) strategies, such as nicMSlesions. However, since the software is trained on fairly homogenous training data, we aimed to test the performance of nicMSlesions in an independent dataset with manual and other automatic lesion segmentations to determine whether this method is suitable for larger, multi-center studies. Manual lesion segmentation was performed in fourteen subjects with MS on sagittal 3D FLAIR images from a 3T GE whole-body scanner with 8-channel head coil. We compared five different categories of automated lesion segmentation methods for their volumetric and spatial agreement with manual segmentation: (i) unsupervised, untrained (LesionTOADS); (ii) supervised, untrained (LST-LPA and nicMSlesions with default settings); (iii) supervised, untrained with threshold adjustment (LST-LPA optimized for current data); (iv) supervised, trained with leave-one-out cross-validation on fourteen subjects with MS (nicMSlesions and BIANCA); and (v) supervised, trained on a single subject with MS (nicMSlesions). Volumetric accuracy was determined by the intra-class correlation coefficient (ICC) and spatial accuracy by Dice's similarity index (SI). Volumes and SI were compared between methods using repeated measures ANOVA or Friedman tests with post-hoc pairwise comparison. The best volumetric and spatial agreement with manual was obtained with the supervised and trained methods nicMSlesions and BIANCA (ICC absolute agreement > 0.968 and median SI > 0.643) and the worst with the unsupervised, untrained method LesionTOADS (ICC absolute agreement = 0.140 and median SI = 0.444). Agreement with manual in the single-subject network training of nicMSlesions was poor for input with low lesion volumes (i.e. two subjects with lesion volumes ≤ 3.0 ml). For the other twelve subjects, ICC varied from 0.593 to 0.973 and median SI varied from 0.535 to 0.606. In all cases, t