{"type":"ScholarlyArticle","title":"Sparse ProtoPatient: Interactive Multi-Prototype Explanations for Clinical Diagnosis Prediction","authors":[{"name":"Conor Fallon","slug":"conor-fallon","given":"Conor","family":"Fallon","affiliation":""},{"name":"Bogdan Kostić","slug":"bogdan-kostic","given":"Bogdan","family":"Kostić","affiliation":""},{"name":"Betty van Aken","slug":"betty-van-aken","given":"Betty","family":"van Aken","affiliation":""},{"name":"Jens-Michalis Papaioannou","slug":"jens-michalis-papaioannou","given":"Jens-Michalis","family":"Papaioannou","affiliation":""},{"name":"Alexei Figueroa","slug":"alexei-figueroa","given":"Alexei","family":"Figueroa","affiliation":""},{"name":"Keno Bressem","slug":"keno-bressem","given":"Keno","family":"Bressem","affiliation":""},{"name":"Alexander Löser","slug":"alexander-loser","given":"Alexander","family":"Löser","affiliation":""}],"abstract":"We present the Sparse ProtoPatient Demo, a publicly available interactive system for interpretable ICD-10 diagnosis prediction from clinical admission notes. \r\nThe system is designed for clinicians in training, researchers, and educators exploring prototype-based diagnostic reasoning.\r\n\r\nThe demo links predictions to learned prototypical patient representations and token-level evidence, allowing users to input custom text or select preset cases, inspect predicted ICD-10 codes, visualize label-wise saliency, retrieve supporting prototype notes, and compare alternative prototype cohorts.\r\nThe demo provides a reproducible platform for interactive inspection of prototype-based clinical reasoning, enabling complementary opinion exploration, model auditing, and teaching of interpretable diagnosis prediction.\r\nThe deployed model is trained on the publicly released CodiEsp corpus (1000 clinical notes, 955 ICD-10 labels) using a sparse multi-prototype architecture with five prototypes per label. \r\nWe use the official machine-translated English CodiEsp-MT release to support English-language interaction.\r\nIt achieves a macro-AUROC of 0.92 on a held-out test set and supports real-time interaction (300ms per query). \r\nThe system is fully containerized for public research and educational use.","doi":"10.24963/ijcai.2026/963","url":"https://ijcai.org/proceedings/2026/963","pdf":"https://ijcai.org/proceedings/2026/0963.pdf","markdown":"https://ijcai.org/proceedings/2026/963.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/963","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Demo Track","track":"Demo Track","number":963,"pages":{"first":8389,"last":8392},"keywords":["AI: Humans and AI","AI: Multidisciplinary Topics and Applications","AI: AI Ethics, Trust, Fairnes","AI: Natural Language Processing"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}