Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

Classification of autoimmune diseases from Peripheral blood TCR repertoires by multimodal multi-instance learning

Created by
  • Haebom

Author

Ruihao Zhang, Mao chen, Fei Ye, Dandan Meng, Yixuan Huang, Xiao Liu

Outline

We developed a multi-instance deep learning framework called EAMil to diagnose systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) with high accuracy using T cell receptor (TCR) sequencing data. By integrating PrimeSeq feature extraction, ESMonehot encoding, and an improved gated attention mechanism, it achieved state-of-the-art performance, achieving AUC of 98.95% for SLE and 97.76% for RA. EAMil successfully identified disease-related genes with over 90% agreement with conventional differential analysis and effectively distinguished disease-specific TCR genes. It stratified disease severity and lesion sites in SLE patients using the SLEDAI score, and showed robustness in classifying multiple disease categories while effectively controlling for confounding factors such as age and gender. This interpretable immune receptor analysis framework provides new insights into autoimmune disease detection and classification with broad potential clinical applications across the spectrum of immune-mediated diseases.

Takeaways, Limitations

•
Takeaways:
◦
Achieved state-of-the-art accuracy (AUC 98.95% and 97.76%) in the diagnosis of SLE and RA.
◦
Successfully identified disease-associated genes and effectively distinguished disease-specific TCR genes.
◦
Severity of disease and site of damage can be diagnosed using the SLEDAI score.
◦
Effectively controls for confounding factors such as age and gender.
◦
Interpretable models provide new insights into the diagnosis and classification of immune-mediated diseases.
◦
Broad clinical applicability to various immune-mediated diseases.
•
Limitations:
◦
Currently, no specific Limitations is explicitly mentioned in the paper. Additional research is expected to be needed in the future to investigate the generalization performance of the model, its extendibility to other autoimmune diseases, and clinical validation.
👍
Made with Slashpage