TY  - JOUR
AU  - Adler, Tim
AU  - Nölke, Jan-Hinrich
AU  - Reinke, Annika
AU  - Tizabi, Minu Dietlinde
AU  - Gruber, Sebastian
AU  - Trofimova, Dasha
AU  - Ardizzone, Lynton
AU  - Jaeger, Paul F
AU  - Buettner, Florian
AU  - Köthe, Ullrich
AU  - Maier-Hein, Lena
TI  - Application-driven validation of posteriors in inverse problems.
JO  - Medical image analysis
VL  - 101
SN  - 1361-8415
CY  - Amsterdam [u.a.]
PB  - Elsevier Science
M1  - DKFZ-2025-00273
SP  - 103474
PY  - 2025
N1  - #EA:E130#LA:E130#
AB  - Current deep learning-based solutions for image analysis tasks are commonly incapable of handling problems to which multiple different plausible solutions exist. In response, posterior-based methods such as conditional Diffusion Models and Invertible Neural Networks have emerged; however, their translation is hampered by a lack of research on adequate validation. In other words, the way progress is measured often does not reflect the needs of the driving practical application. Closing this gap in the literature, we present the first systematic framework for the application-driven validation of posterior-based methods in inverse problems. As a methodological novelty, it adopts key principles from the field of object detection validation, which has a long history of addressing the question of how to locate and match multiple object instances in an image. Treating modes as instances enables us to perform mode-centric validation, using well-interpretable metrics from the application perspective. We demonstrate the value of our framework through instantiations for a synthetic toy example and two medical vision use cases: pose estimation in surgery and imaging-based quantification of functional tissue parameters for diagnostics. Our framework offers key advantages over common approaches to posterior validation in all three examples and could thus revolutionize performance assessment in inverse problems.
KW  - Deep learning (Other)
KW  - Inverse problems (Other)
KW  - Metrics (Other)
KW  - Posterior (Other)
KW  - Validation (Other)
LB  - PUB:(DE-HGF)16
C6  - pmid:39892221
DO  - DOI:10.1016/j.media.2025.103474
UR  - https://inrepo02.dkfz.de/record/298417
ER  -