Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz
Navigation/Suche
- Forscher*innen
- Institutionen
- Projekte
- Publikationen
- Partner
- Geldgeber
- Ausstattung
- Knowhow
- Forschungsfelder
"Köpfe" der aktuellen Meldungen:
Gewählte Publikation:
SHR Neuro Krebs Kardio Lipid Stoffw Microb
Schwitzkowski, M; Kumar, Veeranki, SP; Seidel, BN; Kindle, G; Rusch, S; Kramer, D; Seidel, MG, , ESID, Registry, Working, Party.
Machine learning-assisted diagnosis classification of primary immune dysregulation using IDDA2.1 phenotype profiling.
J Allergy Clin Immunol. 2025;
Doi: 10.1016/j.jaci.2025.10.022
PubMed
FullText
FullText_MUG
- Autor*innen der Med Uni Graz:
- Seidel Markus
- Altmetrics:
- Dimensions Citations:
- Plum Analytics:
- Scite (citation analytics):
- Abstract:
- BACKGROUND: Immune dysregulation, including autoimmunity, autoinflammation, allergy, and malignancy predisposition, adds significant disease burden in primary immune disorders (PID) and inborn errors of immunity (IEI). OBJECTIVE: To evaluate whether the 5-graded immune deficiency and dysregulation activity (IDDA2.1) score, encompassing 21 organ involvement and disease burden parameters, supports diagnosis across a wide spectrum of IEI. PATIENTS AND METHODS: From April 2022 to November 2024, collaborators from 84 centers collected 1,043 IDDA score datasets from 825 patients across 89 IEI (17 disorders with ≥10 patients each; range 1-196/IEI), including 177 scores from 141 treated patients. Supervised machine learning models (k-nearest neighbors, support vector machine, logistic regression, random forest) classified patients into disease groups and ranked corresponding predictive features, while unsupervised UMAP visualized disease-specific clustering. RESULTS: Feature analysis reflected clinicians' recognition of IEI patterns and confirmed internal IDDA score consistency. Phenotype profiles in treated patients remained informative, inversely reflecting anticipated treatment-dependent phenotype amelioration. UMAP effectively distinguished IEI by IDDA2.1 profiles. Genetic disorder prediction achieved 73% overall accuracy, 70% for the correct monogenic IEI, and 93% within the top three predictions; classification reached 43% for IEI-IUIS categories and 59% for 12 "cardinal" IEI (25 genes). CONCLUSIONS: Random Forest Feature importance analysis can inform targeted clinical screening for key disease manifestations. The "top-three" prediction approach demonstrates diagnostic potential, but improved accuracy will require larger, globally shared datasets. Limited sample sizes for rare diseases highlight the necessity of broader collaboration to enhance AI-assisted clinical decision-making in the future.