Artificial intelligence and machine learning in axial spondyloarthritis
Lisa C Adams1, Keno K Bressem2, Denis Poddubnyy3,4
1Department of Diagnostic and Interventional Radiology, Faculty of Medicine.
Current Opinion in Rheumatology
|March 27, 2024
Summary
Artificial intelligence (AI) shows promise in diagnosing axial spondyloarthritis (axSpA) using medical imaging and predictive modeling. Further validation in prospective trials is needed for clinical integration and improved patient care.
Area of Science:
- Rheumatology
- Medical Imaging
- Artificial Intelligence
Background:
- Axial spondyloarthritis (axSpA) diagnosis and management can be challenging.
- Current diagnostic and management strategies may benefit from advanced computational tools.
Purpose of the Study:
- To review the applications and future prospects of artificial intelligence (AI) and machine learning (ML) in axSpA.
- To focus on AI's role in medical imaging, predictive modeling, and patient monitoring for axSpA.
Main Methods:
- Review of current literature on AI and ML in axSpA.
- Analysis of AI applications in medical imaging (X-ray, CT, MRI).
- Evaluation of AI in predictive modeling for disease progression and treatment response.
Main Results:
- AI, especially deep learning, shows potential in diagnosing axSpA by analyzing medical images, sometimes matching or exceeding radiologist performance.
- AI is being utilized for predictive modeling of axSpA progression, personalized treatment, risk assessment, and subtype identification.
- Limitations include variable study designs, small sample sizes, and a predominance of retrospective, single-center studies, hindering generalizability.
Conclusions:
- AI technologies offer significant potential for more accurate, efficient, and personalized axSpA diagnosis and treatment.
- Clinical integration requires rigorous validation, ethical considerations, and healthcare professional training.
- Future AI advancements could enhance clinical expertise and patient care, contingent on prospective multicenter validation and ethical implementation.
