MRI-based classification of neuropsychiatric systemic lupus erythematosus patients with self-supervised contrastive learning

Francesca Inglese, Min‐Seon Kim, Gerda M. Steup‐Beekman, T. Huizinga, Mark A. van Buchem, Jeroen de Bresser, Dae‐Shik Kim, Itamar Ronen

Abstract

Systemic lupus erythematosus (SLE) is a chronic auto-immune disease with a broad spectrum of clinical presentations, including heterogeneous and uncommon neuropsychiatric (NP) syndromes. Accurate diagnosis of neuropsychiatric SLE (NPSLE) is challenging due to a lack of clinically useful biomarkers. This study applies self-supervised contrastive learning on T1-weighted images from a well-defined cohort of 163 SLE patients to distinguish between NPSLE and non-NPSLE cases. Using a 3D convolutional encoder and a linear projector trained with the NT-xent loss, the model achieved 79-83% accuracy. Grad-CAM analysis identified common features in the NPSLE cohort involving the ventricles and periventricular white matter, suggesting a combination of tissue loss and local atrophy.

Introduction: The Clinical-Radiological Paradox

Systemic Lupus Erythematosus (SLE) is a complex autoimmune disease that can affect nearly any organ, including the central nervous system—a condition known as Neuropsychiatric SLE (NPSLE). Diagnosing NPSLE remains a significant clinical challenge. Patients often present with non-specific symptoms such as headaches, seizures, or anxiety. While MRI scans frequently reveal brain abnormalities, these do not always correlate directly with clinical symptoms, a phenomenon termed the "clinical-radiological paradox." Furthermore, distinguishing true NPSLE from "non-NPSLE" (where symptoms arise from other factors like infections or medication) is difficult due to the lack of robust biomarkers.

Traditional supervised deep learning models typically require thousands of labeled images, which are unavailable for rare conditions like NPSLE. This research explores whether self-supervised contrastive learning can learn meaningful brain features from a limited dataset without requiring massive amounts of human-labeled data.

Methodology

The researchers implemented a self-supervised contrastive learning pipeline based on the SimCLR framework, adapted for 3D medical imaging.

Model Architecture and Training

The pipeline consists of a convolutional base encoder and a multi-layer perceptron (MLP) projector. The encoder comprises three layers, each featuring 3D convolution, batch normalization, and max pooling. During the training phase, the model learns by comparing different "views" of the same MRI scan. These views are generated through random 3D transformations, such as cropping, flipping, and rotation.

The machine learning pipeline showing the training phase (Encoder + Projector) and the fine-tuning/test phase (Encoder + Linear layer).

The network is trained using the Normalized Temperature-scaled Cross Entropy (NT-xent) loss function, which maximizes the similarity between different augmented versions of the same patient's image in a latent space while distinguishing them from others.

Illustration of the 3D random crop augmentation used to create different views of the T1-weighted MRI data for contrastive learning.

Dataset and Validation

The study utilized a retrospective dataset of 163 SLE patients from the Leiden University Medical Center (LUMC). The cohort included 34 confirmed NPSLE cases and 129 non-NPSLE cases. All 3T T1-weighted MRI scans were registered to the MNI152 standard template. To ensure robust results with a small sample size, the team employed Monte Carlo cross-validation with six repeated random sub-samplings.

Results and Key Insights

The self-supervised model demonstrated strong performance in classifying NPSLE vs. non-NPSLE:

  • Accuracy: Across the six trials, the model achieved a classification accuracy between 79% and 83%.
  • Recall and Precision: The model showed high recall (~83%), effectively identifying true NPSLE cases. However, precision was lower (~51%), indicating a higher rate of false positives in the non-NPSLE group.
  • Feature Commonality: A critical finding from the Grad-CAM (Gradient-weighted Class Activation Mapping) analysis was that the model identified common structural features across the NPSLE group. In contrast, no such commonality was found for the non-NPSLE group, suggesting that NPSLE has a distinct structural "signature" while non-NPSLE symptoms stem from a heterogeneous mix of causes.

Discussion and Conclusion

The model consistently highlighted regions including the lateral and third ventricles, periventricular white matter, and basal cisterns. These findings suggest that the "signature" of NPSLE involves a combination of tissue loss, local atrophy, and periventricular white matter lesions, which appear hypointense on T1-weighted images.

While promising, the study has limitations. The "gold standard" labels were based on expert consensus, which remains somewhat subjective. Additionally, the use of a single modality (T1-weighted MRI) may limit precision. Future work could incorporate other modalities like FLAIR or Diffusion Tensor Imaging (DTI) to better capture white matter hyperintensities and improve diagnostic accuracy. Overall, this study demonstrates that self-supervised contrastive learning is a powerful tool for capturing diagnostic features in rare diseases with limited datasets.

Summary updated

Abstract

A bstract Introduction/Purpose Systemic lupus erythematosus (SLE) is a chronic auto-immune disease with a broad spectrum of clinical presentations, including heterogeneous and uncommon neuropsychiatric (NP) syndromes. Accurate diagnosis of neuropsychiatric SLE (NPSLE) is challenging due to lack of clinically useful biomarkers. Despite structural brain abnormalities on MRI in NPSLE being a common finding, a robust link between structural abnormalities and NPSLE has not been established, thus their contribution to the distinction between NPSLE patients and patients in which the NP symptoms are not primarily attributed to SLE is limited. Self-supervised contrastive learning algorithms do not require labels, and have been shown to be useful in classification tasks in rare diseases with limited number of datasets. The aim of our study was to apply self-supervised contrastive learning on T 1 -weighted images acquired from a well-defined cohort of SLE patients to distinguish between SLE patients with NP symptoms due to the disease (NPSLE) or and SLE patients with similar symptoms due to other causes (non-NPSLE). Subjects and Methods 163 patients were included. We used 3T MRI T 1 -weighted images registered to the MNI152 template. The training set comprised 68 non-NPSLE and 34 NPSLE patients. During the training procedure, we applied random geometric transformations (cropping, left-right flipping and rotations) between iterations to enrich our data sets. Our 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. We trained the encoder and projector with the Normalized Temperature-scaled Cross Entropy Loss (NT-xent) loss function. We performed a Monte Carlo validation that consisted of 6 repeated random sub-samplings each using a random selection of a small group of samples from each group. Results In the 6 trials described above, 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 the NPSLE population, Gradient-weighted Class Activation Maps (Grad-CAM) were examined voxel-wise. Thresholded Grad-CAM maps show areas of common features identified for the NPSLE cohort, with no such communality found for the non-NPSLE group. Discussion/conclusion The self-supervised contrastive learning model was effective in capturing diagnostic 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 points to 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 T 1 -weighted images.
BibTeX
@inproceedings{inglese2022mri,
  author = {Francesca Inglese and Min‐Seon Kim and Gerda M. Steup‐Beekman and T. Huizinga and Mark A. van Buchem and Jeroen de Bresser and Dae‐Shik Kim and Itamar Ronen},
  title = {{MRI-based classification of neuropsychiatric systemic lupus erythematosus patients with self-supervised contrastive learning}},
  booktitle = {Frontiers in Neuroscience},
  year = {2022},
  doi = {10.3389/fnins.2022.695888},
}
Figure