Towards long-tailed, multi-label disease classification from chest X-ray: Overview of the CXR-LT challenge
0Citations signalées — pas une note de qualité
18Institutions déclarées
6Pays d’affiliation déclarés
Résumé fourni par la source
, on long-tailed, multi-label thorax disease classification from chest X-rays (CXRs). We publicly release a large-scale benchmark dataset of over 350,000 CXRs, each labeled with at least one of 26 clinical findings following a long-tailed distribution. We synthesize common themes of top-performing solutions, providing practical recommendations for long-tailed, multi-label medical image classification. Finally, we use these insights to propose a path forward involving vision-language foundation models for few- and zero-shot disease classification.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.
Contrôle bibliographique ouvert
La source scientifique ouverte est momentanément indisponible.
Institutions déclarées
The University of Texas at AustinCornell UniversityCarnegie Mellon UniversityMassachusetts Institute of TechnologyHo Chi Minh City University of ScienceKorea Telecom (South Korea)Ajou UniversityShanghai Jiao Tong UniversityShandong Jiaotong UniversityCentre for Artificial Intelligence and RoboticsThe University of TokyoSK Group (South Korea)Seoul National UniversityBeth Israel Deaconess Medical CenterHarvard UniversityCancer Research And BiostatisticsNational Center for Biotechnology InformationNational Institutes of Health Clinical Center
Une affiliation ne permet pas de déduire la nationalité d’un auteur.
Sujets associés
COVID-19 diagnosis using AIMachine Learning in HealthcareTuberculosis Research and Epidemiology