Benutzer: Gast  Login
Titel:

Improved Post-hoc Probability Calibration for Out-of-Domain MRI Segmentation

Dokumenttyp:
Proceedings Paper
Autor(en):
Ouyang, Cheng; Wang, Shuo; Chen, Chen; Li, Zeju; Bai, Wenjia; Kainz, Bernhard; Rueckert, Daniel
Abstract:
Probability calibration for deep models is highly desirable in safety-critical applications such as medical imaging. It makes output probabilities of deep networks interpretable, by aligning prediction probability with the actual accuracy in test data. In image segmentation, well-calibrated probabilities allow radiologists to identify regions where model-predicted segmentations are unreliable. These unreliable predictions often occur to out-of-domain (OOD) images that are caused by imaging artif...     »
Zeitschriftentitel:
Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv
Jahr:
2022
Band / Volume:
13563
Seitenangaben Beitrag:
59-69
Volltext / DOI:
doi:10.1007/978-3-031-16749-2_6
Print-ISSN:
0302-9743
TUM Einrichtung:
Institut für KI und Informatik in der Medizin (Prof. Rückert)
 BibTeX