Academic Journal
Rare-Disease Diagnosis on the ZebraMap Multimodal Case Report Dataset: A Hybrid Pipeline with Grounded Explainability.
| 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. |
| References: | Diagnostics (Basel). 2025 Dec 29;16(1):. (PMID: 41515602) NPJ Digit Med. 2025 Oct 7;8(1):600. (PMID: 41057566) BMC Med Inform Decis Mak. 2024 Oct 8;24(1):289. (PMID: 39375687) J Biomed Inform. 2024 Sep;157:104702. (PMID: 39084480) Orphanet J Rare Dis. 2020 Jun 9;15(1):145. (PMID: 32517778) Pediatr Res. 2021 Jan;89(2):295-300. (PMID: 32932427) Eur J Hum Genet. 2020 Feb;28(2):165-173. (PMID: 31527858) Orphanet J Rare Dis. 2020 Sep 24;15(1):263. (PMID: 32972444) Orphanet J Rare Dis. 2020 Apr 16;15(1):94. (PMID: 32299466) Am J Hum Genet. 2020 Sep 3;107(3):403-417. (PMID: 32755546) Orphanet J Rare Dis. 2022 Jun 18;17(1):233. (PMID: 35717227) Am J Hum Genet. 2009 Oct;85(4):457-64. (PMID: 19800049) Lancet. 2008 Jun 14;371(9629):2039-41. (PMID: 18555915) Nature. 2023 Aug;620(7972):172-180. (PMID: 37438534) BMC Med Inform Decis Mak. 2023 Feb 2;23(1):26. (PMID: 36732730) Bioinformatics. 2023 Nov 1;39(11):. (PMID: 37930897) J Am Med Inform Assoc. 2024 Apr 3;31(4):968-974. (PMID: 38383050) Nat Rev Drug Discov. 2020 Feb;19(2):77-78. (PMID: 32020066) Lancet Digit Health. 2026 Feb;8(2):100953. (PMID: 41748380) Biomed Eng Lett. 2025 Aug 22;15(5):845-863. (PMID: 40917142) |
| 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: 20260813 |
| Update Code: | 20260813 |
| PubMed Central ID: | PMC13259397 |
| DOI: | 10.3390/s26113582 |
| PMID: | 42281096 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1424-8220 |
|---|---|
| DOI: | 10.3390/s26113582 |