Breakthrough in Arthritis Diagnosis: Fuzzy Logic and Machine Learning Collaboration

Strategic Insights -

Researchers from Sahand University of Technology, led by Mohammed Fadhil Mahdi, have proposed a novel combination of explainable machine learning and fuzzy evaluation frameworks to improve the diagnostic performance and interpretation of rheumatic and autoimmune diseases. This innovative approach addresses the challenges of overlapping symptoms, complex clinical presentations, and the lack of interpretability in traditional machine learning models. The study utilized a dataset of 12,085 patients collected from Iraq's hospitals and health centers between 2019 and 2024, which includes 14 features in seven classes.

Key Takeaways:

  • The researchers proposed a new combination of explainable machine learning and fuzzy evaluation frameworks to improve the diagnostic performance and interpretation of rheumatic and autoimmune diseases.
  • The study addressed three major challenges: overlapping symptoms, complex clinical presentations, and the lack of interpretability in traditional machine learning models.
  • The dataset used in the study consisted of 12,085 patients and included 14 features in seven classes (rheumatoid arthritis, reactive arthritis, ankylosing spondylitis, Sjogren syndrome, systemic lupus erythematosus, psoriatic arthritis, and normal).
  • The researchers applied fuzzy decision by opinion score method (FDOSM) to select the optimal model, which involved assessments from three domain experts.
  • The explainable artificial intelligence (XAI) technique provided global and local explanations for model predictions, increasing transparency and reliability in clinical decision-making.
  • The results showed that the FDOSM yielded gradient boosting as the best model with an accuracy of 86.89%, precision of 87.35%, and kappa of 84.51%.
  • The study concluded that the proposed approach can increase confidence and trustworthiness in clinical decision-making and healthcare applications.

Statistics:

  • 12,085 patients were included in the dataset.
  • 14 features and 7 classes were included in the dataset.
  • The proposed approach consisted of explainable machine learning and fuzzy evaluation frameworks.
  • The FDOSM yielded gradient boosting as the best model with an accuracy of 86.89%, precision of 87.35%, and kappa of 84.51%.
  • The study evaluated the performance of 12 different machine learning models.

Sources:

  • Fuzzy evaluation and explainable machine learning for diagnosis of rheumatic and autoimmune diseases. PeerJ Computer Science, 2025,11():e3096.
  • NewsRx. Sahand University of Technology Researchers Focus on Arthritis (Fuzzy evaluation and explainable machine learning for diagnosis of rheumatic and autoimmune diseases). Health & Medicine Week. August 29, 2025; p 7019.