MRI-Based Classification of Neuropsychiatric Systemic Lupus Erythematosus Patients With Self-Supervised Contrastive Learning

Abstract

Introduction/PurposeSystemic lupus erythematosus (SLE) is a chronic auto-immune disease with a broad spectrum of clinical presentations, including heterogeneous neuropsychiatric (NP) syndromes. Structural brain abnormalities are commonly found in SLE and NPSLE, but their role in diagnosis is limited, and their usefulness in distinguishing between NPSLE patients and patients in which the NP symptoms are not primarily attributed to SLE (non-NPSLE) is non-existent. Self-supervised contrastive learning algorithms proved to be useful in classification tasks in rare diseases with limited number of datasets. Our aim was to apply self-supervised contrastive learning on T1-weighted images acquired from a well-defined cohort of SLE patients, aiming to distinguish between NPSLE and non-NPSLE patients.Subjects and MethodsWe used 3T MRI T1-weighted images of 163 patients. The training set comprised 68 non-NPSLE and 34 NPSLE patients. We applied random geometric transformations between iterations to augment our data sets. The ML pipeline consisted of convolutional base encoder and linear projector. To test the classification task, the projector was removed and one linear layer was measured. Validation of the method consisted of 6 repeated random sub-samplings, each using a random selection of a small group of patients of both subtypes.ResultsIn the 6 trials, between 79% and 83% of the patients were correctly classified as NPSLE or non-NPSLE. For a qualitative evaluation of spatial distribution of the common features found in both groups, Gradient-weighted Class Activation Maps (Grad-CAM) were examined. Thresholded Grad-CAM maps show areas of common features identified for the NPSLE cohort, while no such communality was found for the non-NPSLE group.Discussion/ConclusionThe self-supervised contrastive learning model was effective in capturing common brain MRI features from a limited but well-defined cohort of SLE patients with NP symptoms. The interpretation of the Grad-CAM results is not straightforward, but indicates involvement of the lateral and third ventricles, periventricular white matter and basal cisterns. We believe that the common features found in the NPSLE population in this study indicate a combination of tissue loss, local atrophy and to some extent that of periventricular white matter lesions, which are commonly found in NPSLE patients and appear hypointense on T1-weighted images.

Serotonin Syndrome-Mimicking Manifestations in a Patient with Systemic Lupus Erythematosus

“…In addition, image techniques and functional anatomy of the nervous system may be applied for the classification of NPSLE. New image analysis techniques such as self-supervised contrastive learning, volumetric measures of globus pallidus, or quantitative susceptibility mapping as well as activation pattern assessed by functional MRI may detect subtle serial changes in normal-appearing areas and provide new insight into its pathogenic pathway [ 6 , 24 , 25 , 26 ], which hopefully will lead to the improvement in diagnosis and management.…”

“…The association between anti-ribosomal P antibody and the presence of neuropsychiatric lupus (NPSLE) has been well-established, but the pathogenic role of the autoantibody remains unclear [ 2 , 3 , 4 ]. Neuroimage protocols of magnetic resonance image (MRI) are used extensively to define the anatomic basis of NPSLE [ 5 , 6 , 7 , 8 ].…”

Systemic lupus in the era of machine learning medicine

“…Five reports used convolutional neural networks (CNN) 45 on image data with topics ranging from NPSLE diagnosis from MRI images, 46 diagnosis of SLE retinopathy from funduscopic images, 23 diagnosis of cutaneous lupus from lesion images, 47 segmentation of staining from lupus nephritis (LN) pathology images 48 and segmentation of glomeruli on LN biopsy images. 43 Three of the reports used a deep learning technique called Grad-CAM 49 that identifies the region of an image that will contribute the most to the final model.…”

A Review of Predictive and Contrastive Self-supervised Learning for Medical Images

“…The proposed model demonstrated high-performance cephalometric landmark detection, comparable to popular fully supervised approaches utilizing more than one training image. Ali et al [199] used 3D SimCLR during pretraining and Monte Carlo dropout during prediction on two tasks, including 3D CT pancreas tumour and 3D MRI brain tumour segmentation. Inglese et al [200] followed a similar optimization method of SimCLR to train an SSL network for distinguishing between two diagnostically different systemic lupus erythematosus patient groups.…”