Year: 2026
Type: article
Abstract: The integration of multi-stain histopathology images through deep learning poses a significant challenge. Current approaches struggle with data heterogeneity and missing data, as concatenating multi-stain features may not effectively model stain-specific and cross-stain interactions. We introduce UNICORN (UNiversal stain Integration network for CORonary classificatioN), a two-stage, end-to-end trainable model comprising transformer self-attention blocks to process multi-stain histopathology for atherosclerosis severity prediction. The initial stage employs domain-specific expert models to extract features from each staining. An aggregation expert model then integrates features by learning their interactions. On a multi-class, multi-stain whole slide images (WSIs) dataset of atherosclerotic lesions from Munich Cardiovascular Studies Biobank (MISSION), UNICORN achieved a classification accuracy of 0.68, significantly outperforming state-of-the-art models. UNICORN identifies relevant tissue phenotypes across stainings and implicitly models disease progression. Its explainability and effectiveness in predicting atherosclerosis progression highlight the potential for broader applications in medical research and decision support.
Source: npj Digital Medicine
Authors: Valentin Koch, Sabine Bauer, Shweta Mahajan, Valerio Lupperger, Michael Joner, Heribert Schunkert, Julia A. Schnabel, Moritz von Scheidt, Carsten Marr
Institutions: Helmholtz Munich, Technical University of Munich, TUM Klinikum, German Centre for Cardiovascular Research, Deutsches Herzzentrum München, Munich Leukemia Laboratory (Germany), King's College London, LMU Klinikum, Deutsches Konsortium für Translationale Krebsforschung, Ludwig-Maximilians-Universität München
Language: English
Topic: AI in cancer detection
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $4,090
Funders: Foundation for Cardiovascular Research, Deutsches Zentrum für Herz-Kreislaufforschung, Corona-Stiftung, British Heart Foundation, European Commission, Deutsche Forschungsgemeinschaft, Fondation Leducq, Bundesministerium für Bildung und Forschung, Deutsche Herzstiftung, German Network for Bioinformatics Infrastructure, Horizon 2020
Awards: G1147967356, G1159569226, G1254183670, G1408866189, G1511448630, de.NBI - Etablierungsphase - Heidelberg Center for Human Bioinformatics - HD-HuB, G2082019117, de.NBI - Etablierungsphase - Leistungszentrum - BiGi - Bioinformatisches Resourcenzentrum für mikrobielle Genomforschung in Biotechnologie und Medizin, G2502478445, G2511594427, G2813928278, G313332700, G431688250, de.NBI - Etablierungsphase - Leistungszentrum CiBi - Zentrum für integrative Bioinformatik, G5121333388, G5139259834, de.NBI - Etablierungsphase - Heidelberg Center for Human Bioinformatics - HD-HuB - Humane Genetik und Genomik, Humane Mikrobiomik, systematische Phänotypisierung humaner Zellen, G524245738, G5576285491, de.NBI - Etablierungsphase - Leistungszentrum: RBC - RNA Bioinformatic, de.NBI - Etablierungsphase - Heidelberg Center for Human Bioinformatics - HD-HuB - Humane Genetik und Genomik, Humane Metagenomik, systematische Phänotypisierung humaner Zellen, Epigenetik und de.NBI Cloud Standort Heidelberg/Berlin, G627612317, G6645136632, de.NBI - Etablierungsphase - Leistungszentrum - BioInfra.Prot - Bioinformatik der Proteomik, G7502528521, G7614570622, G7636979992, de.NBI - Etablierungsphase - Koordinierungs- und Administrationseinheit - CAU - Geschäftsstelle, G8205150067, de.NBI - Etablierungsphase - Leistungszentrum - BiGi - Bioinformatisches Resourcenzentrum für mikrobielle Genomforschung in Biotechnologie und Medizin, G8424115693, de.NBI - Etablierungsphase-European Molecular Biology Laboratory-HD-HUB
OpenAlex ID: W4403796237