AI for Allergic Rhinitis — Clinical Decision Support for Etiologic Diagnosis
OVERVIEWThe @IT-2020 project (PubMed 41903764) developed a modular, AI-driven clinical decision support system (CDSS) for etiologic diagnosis of seasonal allergic rhinitis (SAR) — a clinically challenging task in temperate climates where patients are often polysensitized to multiple overlapping pollens. Three expert raters trained ML models by identifying culprit pollens in 100 SAR patients in Rome, Italy, using a Delphi-like consensus process. The CDSS includes three progressive diagnostic modules: (a) clinical history and skin prick test (SPT), (b) adding molecular specific IgE, and (c) adding electronic/environmental diary data. Best-performing ML models for each module reliably replicated expert diagnosis, achieving area under the ROC curve >95%. SHAP analysis confirmed interpretability, showing that clinical features and sensitization patterns contributed meaningfully to predictions. The system demonstrated geographical generalizability (consistent performance in 92 Albanian patients from Tirana), temporal generalizability (maintained high performance with only 45 days of monitoring), and AIT prescription adaptability (reproducing gold-standard prescriptions). In a diagnostic challenge, the CDSS outperformed 24 physicians. The study represents a proof-of-concept for AI-assisted precision diagnosis in allergology, with implications for enabling personalized allergen immunotherapy prescriptions. The modular architecture allows the CDSS to adapt to patient complexity and available…
Subscribe to read full wiki analysis
Full wiki content — including source provenance, contradiction tracking, related pages, and analyst notes — is available to Subscriber members.
View Plans & SubscribeNew users get a 7-day trial — no credit card required.