Academic Journal

Rare-Disease Diagnosis on the ZebraMap Multimodal Case Report Dataset: A Hybrid Pipeline with Grounded Explainability.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Rare-Disease Diagnosis on the ZebraMap Multimodal Case Report Dataset: A Hybrid Pipeline with Grounded Explainability.
Συγγραφείς: Islam MS; Department of Computer Science, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 21589, Saudi Arabia.; Department of Computer Science, Faculty of Science and Information Technology, Daffodil International University, Birulia 1216, Bangladesh., Jamal A; Department of Computer Science, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 21589, Saudi Arabia.; Center of Research Excellence in Artificial Intelligence and Data Science, King Abdulaziz University, Jeddah 21589, Saudi Arabia., Alkhathlan A; Department of Computer Science, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 21589, Saudi Arabia.
Πηγή: Sensors (Basel, Switzerland) [Sensors (Basel)] 2026 Jun 04; Vol. 26 (11). Date of Electronic Publication: 2026 Jun 04.
Τύπος έκδοσης: Journal Article
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: MDPI Country of Publication: Switzerland NLM ID: 101204366 Publication Model: Electronic Cited Medium: Internet ISSN: 1424-8220 (Electronic) Linking ISSN: 14248220 NLM ISO Abbreviation: Sensors (Basel) Subsets: MEDLINE
Imprint Name(s): Original Publication: Basel, Switzerland : MDPI, c2000-
Ιατρικοί όροι (MeSH): Rare Diseases*/diagnosis , Case Reports as Topic* , Classification Algorithms* , Large Language Models*, Humans
Περίληψη: Rare-disease diagnosis is difficult because clinicians must identify plausible conditions from a large, severely imbalanced disease space using evidence distributed across clinical narratives, structured findings, and image-linked descriptions. This paper presents a hybrid pipeline with caption-mediated multimodal fusion for ranked rare-disease diagnosis and grounded explanation, developed and evaluated on the ZebraMap multimodal case-report dataset (69,146 structured cases; 1727 diseases). Grouped train-validation-test splitting by source article was applied to prevent leakage, and a sequential pipeline was constructed combining BM25 lexical retrieval, a class-balanced TF-IDF classifier, MedCPT dense retrieval and cross-encoder reranking, caption-based image-aware late fusion, and post hoc grounded explanation generation. The final pipeline achieved test MRR 0.3905 and Recall@10 0.5507 (nDCG@10 0.4273), while the strongest individual component, the class-balanced TF-IDF classifier, reached MRR 0.4200 and Recall@10 0.6279; the hybrid pipeline therefore integrates ranking with grounded explanation rather than maximizing single-metric diagnostic accuracy. On 256 explained cases, the explanation module achieved citation coverage 0.7334 and usefulness 3.8734, exposing a tradeoff between diagnostic accuracy and explanation richness. These results indicate that a hybrid retrieval-and-classification approach can support ranked rare-disease differential diagnosis and that grounded explanation quality can be evaluated quantitatively, extending computational support for the prolonged rare-disease diagnostic process.
Contributed Indexing: Keywords: benchmarking; caption-mediated multimodal fusion; clinical decision support; explainability; information retrieval; large language models; rare-disease diagnosis
Entry Date(s): Date Created: 20260612 Date Completed: 20260613 Latest Revision: 20260726
Update Code: 20260726
PubMed Central ID: PMC13259397
DOI: 10.3390/s26113582
PMID: 42281096
Βάση Δεδομένων: MEDLINE