PADCHEST: Un gran conjunto de datos de imágenes de rayos X de tórax con informes anotados de etiquetas múltiples
Spanish
Spain, Valencia
PadChest
数据描述
PadChest: Un gran conjunto de datos de imágenes de rayos X de tórax con informes anotados de etiquetas múltiples. Presentamos un conjunto de datos etiquetados a gran escala y de alta resolución de radiografías de tórax para la ex-ploración automatizada de imágenes médicas junto con sus informes asociados. Este conjunto de datos incluye más de 160.000 imágenes de 67.000 pacientes que fueron interpretadas e informadas por radiólogos en el Hospital San Juan (España) desde 2009 hasta 2017, cubriendo seis vistas de posición diferentes e información adicional sobre la adquisición de la imagen y la demografía del paciente. Los informes se etiquetaron con 174 hallazgos radiográficos diferentes, 19 diagnósticos diferenciales y 104 localizaciones anatómicas organizadas como una taxonomía jerárquica asignada a la terminología estándar del Sistema de Lenguaje Médico Unificado (UMLS).
bimcv.cipf.es
IP: 195.77.22.54
加载中...
相关论文
6Exploring scalable medical image encoders beyond text supervision
Fernando Pérez-García•Harshita Sharma•Sam Bond-Taylor•Kenza Bouzid•Valentina Salvatelli等 15 人
Nature Machine Intelligence
2025
•2025/1/13
•00 p.1-12
Language-supervised pretraining has proven to be a valuable method for extracting semantically meaningful features from images, serving as a foundational element in multimodal systems within the computer vision and medical imaging domains. However, the computed features are limited by the informatio...
DiagnosisMedical researchRadiography
10.1038/S42256-024-00965-W
ISSN:2522-5839
A generative model uses healthy and diseased image pairs for pixel-level chest X-ray pathology localization
Kaiming Dong•Yuxiao Cheng•Kunlun He•Jinli Suo
Nature Biomedical Engineering
2025
•2025/7/14
•00 p.1-13
Medical artificial intelligence (AI) offers potential for automatic pathological interpretation, but a practicable AI model demands both pixel-level accuracy and high explainability for diagnosis. The construction of such models relies on substantial training data with fine-grained labelling, which ...
Image processingMachine learningRadiography
10.1038/S41551-025-01456-Y
ISSN:2157-846X
Expert-level detection of pathologies from unannotated chest X-ray images via self-supervised learning
Tiu Ekin•Talius Ellie•Patel Pujan•Langlotz Curtis P.•Ng Andrew Y.等 6 人
Nature Biomedical Engineering
2022
•2022/9/15
•Vol.6 No.12 p.1399-1406
In tasks involving the interpretation of medical images, suitably trained machine-learning models often exceed the performance of medical experts. Yet such a high-level of performance typically requires that the models be trained with relevant datasets that have been painstakingly annotated by exper...
Health careMedical imaging
10.1038/S41551-022-00936-9
ISSN:2157-846X
AI for radiographic COVID-19 detection selects shortcuts over signal
Alex J. DeGrave•Joseph D. Janizek•Su-In Lee
Nature Machine Intelligence
2021
•2021/5/31
•00 p.1-10
Artificial intelligence (AI) researchers and radiologists have recently reported AI systems that accurately detect COVID-19 in chest radiographs. However, the robustness of these systems remains unclear. Using state-of-the-art techniques in explainable AI, we demonstrate that recent deep learning sy...
Computational sciencePredictive medicineRadiographySARS-CoV-2
10.1038/S42256-021-00338-7
ISSN:2522-5839
Knowledge-enhanced visual-language pre-training on chest radiology images
Xiaoman Zhang•Chaoyi Wu•Ya Zhang•Weidi Xie•Yanfeng Wang
Nature Communications
2023
•2023/7/28
•Vol.14 No.1 p.1-12
While multi-modal foundation models pre-trained on large-scale data have been successful in natural language understanding and vision recognition, their use in medical domains is still limited due to the fine-grained nature of medical tasks and the high demand for domain knowledge. To address this c...
DiseasesHealth careMedical research
10.1038/S41467-023-40260-7
ISSN:2041-1723
CLEAR: an auditable foundation model for radiology grounded in clinical concepts
Tianyu Han•Riga Wu•Yu Tian•Firas Khader•Lisa C. Adams等 12 人
Nature Biomedical Engineering
2026
•2026/7/22
•00 p.1-16
‘Black box’ deep learning models for medical image interpretation limit clinical trust and analysis of performance degradation. Here we introduce Concept-Level Embeddings for Auditable Radiology (CLEAR), an auditable foundation model based on clinical concepts. Trained on over 0.87 million image–rep...
Computational scienceMedical imaging
10.1038/S41551-026-01741-4
ISSN:2157-846X