Clinical Diagnosis of Rare Diseases Using Leaky Noisy-OR Bayesian Networks.
WHY IT MATTERS
If this diagnostic tool becomes available to patients and doctors, it could significantly reduce the years of diagnostic delay that rare disease patients typically experience by quickly narrowing down which rare diseases match their symptoms.
Researchers created a computer system that helps doctors diagnose rare diseases by analyzing a patient's symptoms. The system uses artificial intelligence to compare a patient's symptoms against information about over 8,000 rare diseases and correctly identified the right diagnosis about 56% of the time when looking at the top three possibilities. This is similar to how other leading symptom-checking systems perform.
Clinical Diagnosis of Rare Diseases Using Leaky Noisy-OR Bayesian Networks. Abstract: This study presents a probabilistic method for the clinical diagnosis of rare diseases using leaky noisy-OR Bayesian networks automatically constructed from Orphanet and Human Phenotype Ontology data. The resulting model represents diseases and phenotypes as binary variables linked by causal probabilities derived from standardized annotations. Loopy belief propagation enables efficient approximate inference of disease posterior probabilities in large networks containing over 8,000 diseases and 9,000 finding variables. Evaluation on real clinical cases achieves 56.2% Top-3 diagnostic accuracy, in line with the reported performance of leading phenotype-based systems. The proposed framework demonstrates that interpretable and knowledge-grounded probabilistic reasoning can achieve state-of-the Authors: Roucoux et al. Journal: Studies in health technology and informatics MeSH: Bayes Theorem, Rare Diseases, Humans, Diagnosis, Computer-Assisted