Influence of Classification Task and Distribution Shift Type on OOD Detection in Fetal Ultrasound
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915507002671104 |
|---|---|
| author | Wong, Chun Kit Christensen, Anders N. Bercea, Cosmin I. Schnabel, Julia A. Tolsgaard, Martin G. Feragen, Aasa |
| author_facet | Wong, Chun Kit Christensen, Anders N. Bercea, Cosmin I. Schnabel, Julia A. Tolsgaard, Martin G. Feragen, Aasa |
| contents | Reliable out-of-distribution (OOD) detection is important for safe deployment of deep learning models in fetal ultrasound amidst heterogeneous image characteristics and clinical settings. OOD detection relies on estimating a classification model's uncertainty, which should increase for OOD samples. While existing research has largely focused on uncertainty quantification methods, this work investigates the impact of the classification task itself. Through experiments with eight uncertainty quantification methods across four classification tasks, we demonstrate that OOD detection performance significantly varies with the task, and that the best task depends on the defined ID-OOD criteria; specifically, whether the OOD sample is due to: i) an image characteristic shift or ii) an anatomical feature shift. Furthermore, we reveal that superior OOD detection does not guarantee optimal abstained prediction, underscoring the necessity to align task selection and uncertainty strategies with the specific downstream application in medical image analysis. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_18326 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Influence of Classification Task and Distribution Shift Type on OOD Detection in Fetal Ultrasound Wong, Chun Kit Christensen, Anders N. Bercea, Cosmin I. Schnabel, Julia A. Tolsgaard, Martin G. Feragen, Aasa Computer Vision and Pattern Recognition Reliable out-of-distribution (OOD) detection is important for safe deployment of deep learning models in fetal ultrasound amidst heterogeneous image characteristics and clinical settings. OOD detection relies on estimating a classification model's uncertainty, which should increase for OOD samples. While existing research has largely focused on uncertainty quantification methods, this work investigates the impact of the classification task itself. Through experiments with eight uncertainty quantification methods across four classification tasks, we demonstrate that OOD detection performance significantly varies with the task, and that the best task depends on the defined ID-OOD criteria; specifically, whether the OOD sample is due to: i) an image characteristic shift or ii) an anatomical feature shift. Furthermore, we reveal that superior OOD detection does not guarantee optimal abstained prediction, underscoring the necessity to align task selection and uncertainty strategies with the specific downstream application in medical image analysis. |
| title | Influence of Classification Task and Distribution Shift Type on OOD Detection in Fetal Ultrasound |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.18326 |