Abstract
BackgroundMagnetic resonance imaging (MRI) has a wide range of applications in medical imaging. Recently, studies based on deep learning algorithms have demonstrated powerful processing capabilities for medical imaging data. Previous studies have mostly focused on common diseases that usually have large scales of datasets and centralized the lesions in the brain. In this paper, we used deep learning models to process MRI images to differentiate the rare neuromyelitis optical spectrum disorder (NMOSD) from multiple sclerosis (MS) automatically, which are characterized by scattered and overlapping lesions.MethodsWe proposed a novel model structure to capture 3D MRI images’ essential information and converted them into lower dimensions. To empirically prove the efficiency of our model, firstly, we used a conventional 3-dimensional (3D) model to classify the T2-weighted fluid-attenuated inversion recovery (T2-FLAIR) images and proved that the traditional 3D convolutional neural network (CNN) models lack the learning capacity to distinguish between NMOSD and MS. Then, we compressed the 3D T2-FLAIR images by a two-view compression block to apply two different depths (18 and 34 layers) of 2D models for disease diagnosis and also applied transfer learning by pre-training our model on ImageNet dataset.ResultsWe found that our models possess superior performance when our models were pre-trained on ImageNet dataset, in which the models’ average accuracies of 34 layers model and 18 layers model were 0.75 and 0.725, sensitivities were 0.707 and 0.708, and specificities were 0.759 and 0.719, respectively. Meanwhile, the traditional 3D CNN models lacked the learning capacity to distinguish between NMOSD and MS.ConclusionThe novel CNN model we proposed could automatically differentiate the rare NMOSD from MS, especially, our model showed better performance than traditional3D CNN models. It indicated that our 3D compressed CNN models are applicable in handling diseases with small-scale datasets and possess overlapping and scattered lesions.
Highlights
Neuromyelitis optical spectrum disorder (NMOSD) is a rare aquaporin-4 immunoglobin G antibody (AQP4-IgG) mediated chronic disorder of the brain and spinal cord (Wingerchuk et al, 2007, 2015)
The 3D model we used was based on the models proposed by Payan and Montana (2015), Hara et al (2017, 2018), Wang H. et al (2019), which were applied traditional 3D ResNet for classification tasks and achieved desired results
Proposed a compression block to map the high dimensional 3D data into a lower dimension. It can extract the information from the overlapping lesion locations by learning each single magnetic resonance imaging (MRI) section and building long-range dependence at the same time
Summary
Neuromyelitis optical spectrum disorder (NMOSD) is a rare aquaporin-4 immunoglobin G antibody (AQP4-IgG) mediated chronic disorder of the brain and spinal cord (Wingerchuk et al, 2007, 2015). Considered a subtype of multiple sclerosis (MS), NMOSD has been recognized as a distinct clinical entity based on unique immunologic features in recent years (Wingerchuk et al, 2015). Up to 70% of NMOSD patients have brain lesions visible on magnetic resonance imaging (MRI) (Kim et al, 2015). Studies based on machine learning to discriminate NMOSD from MS are limited. Studies based on deep learning algorithms have demonstrated powerful processing capabilities for medical imaging data. We used deep learning models to process MRI images to differentiate the rare neuromyelitis optical spectrum disorder (NMOSD) from multiple sclerosis (MS) automatically, which are characterized by scattered and overlapping lesions
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have