Artificial Intelligence in Action: A Comprehensive Review on Machine and Deep Learning Methods in Sjögren's Syndrome...

Area of Science:

• Immunology
• Medical Informatics
• Biomedical Engineering

Background:

• Sjögren's syndrome (SS) is a chronic autoimmune disease with varied symptoms, often leading to delayed diagnosis.
• Current diagnostic methods for SS can be nonspecific and rely on invasive procedures.

Purpose of the Study:

• To review the role of artificial intelligence (AI), machine learning (ML), and deep learning (DL) in improving Sjögren's syndrome diagnosis.
• To highlight AI's potential in enhancing diagnostic accuracy across various clinical settings.

Main Methods:

• Review of AI applications in histopathological analysis of salivary gland biopsies.
• Analysis of DL models in imaging diagnostics like ultrasonography and CT scans.
• Evaluation of noninvasive methods (Raman spectroscopy, tongue imaging) combined with ML.
• Exploration of AI in genomic and metabolomic profiling for biomarker discovery.
• Assessment of ML models for early case identification using electronic health records (EHRs).

Main Results:

• AI models demonstrate high accuracy in automated histopathological analysis, reducing observer variability.
• DL models in imaging surpass inexperienced radiologists in detecting glandular abnormalities.
• Noninvasive techniques show promise as alternatives to traditional SS diagnostics.
• AI aids in identifying novel biomarkers and molecular signatures for SS.
• ML models in primary care show potential for early SS identification and reduced referral delays.

Conclusions:

• AI integration across diverse diagnostic modalities can significantly improve Sjögren's syndrome detection.
• AI facilitates faster, more objective, and accessible diagnostics, even in resource-limited settings.
• AI unifies disparate data sources for timely and personalized SS diagnosis.