Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Rethinking Annotation Granularity for Overcoming Shortcuts in Deep Learning–based Radiograph Diagnosis: A Multicenter Study
24
Zitationen
13
Autoren
2022
Jahr
Abstract
Purpose: To evaluate the ability of fine-grained annotations to overcome shortcut learning in deep learning (DL)-based diagnosis using chest radiographs. Materials and Methods: = 496). Performance was evaluated using receiver operating characteristic (ROC) curve analysis. Results: < .001). Conclusion: © RSNA, 2022.
Ähnliche Arbeiten
La certeza de lo impredecible: Cultura Educación y Sociedad en tiempos de COVID19
2020 · 19.307 Zit.
A Multi-Modal Distributed Real-Time IoT System for Urban Traffic Control (Invited Paper)
2024 · 14.299 Zit.
UNet++: A Nested U-Net Architecture for Medical Image Segmentation
2018 · 8.787 Zit.
Review of deep learning: concepts, CNN architectures, challenges, applications, future directions
2021 · 7.398 Zit.
scikit-image: image processing in Python
2014 · 6.834 Zit.
Autoren
Institutionen
- Chinese University of Hong Kong(HK)
- Chinese Academy of Medical Sciences & Peking Union Medical College(CN)
- Hong Kong University of Science and Technology(HK)
- Chinese University of Hong Kong, Shenzhen(CN)
- University of Hong Kong(HK)
- Shenzhen Luohu People's Hospital(CN)
- Guangdong Academy of Medical Sciences(CN)
- Guangdong Provincial People's Hospital(CN)
- Queen Mary Hospital(CN)
- Hospital Authority(HK)
- Shenzhen Institutes of Advanced Technology(CN)