Aggregated Attributions for Explanatory Analysis of 3D Segmentation Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912137767550976 |
|---|---|
| author | Chrabaszcz, Maciej Baniecki, Hubert Komorowski, Piotr Płotka, Szymon Biecek, Przemyslaw |
| author_facet | Chrabaszcz, Maciej Baniecki, Hubert Komorowski, Piotr Płotka, Szymon Biecek, Przemyslaw |
| contents | Analysis of 3D segmentation models, especially in the context of medical imaging, is often limited to segmentation performance metrics that overlook the crucial aspect of explainability and bias. Currently, effectively explaining these models with saliency maps is challenging due to the high dimensions of input images multiplied by the ever-growing number of segmented class labels. To this end, we introduce Agg^2Exp, a methodology for aggregating fine-grained voxel attributions of the segmentation model's predictions. Unlike classical explanation methods that primarily focus on the local feature attribution, Agg^2Exp enables a more comprehensive global view on the importance of predicted segments in 3D images. Our benchmarking experiments show that gradient-based voxel attributions are more faithful to the model's predictions than perturbation-based explanations. As a concrete use-case, we apply Agg^2Exp to discover knowledge acquired by the Swin UNEt TRansformer model trained on the TotalSegmentator v2 dataset for segmenting anatomical structures in computed tomography medical images. Agg^2Exp facilitates the explanatory analysis of large segmentation models beyond their predictive performance. The source code is publicly available at https://github.com/mi2datalab/agg2exp. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_16653 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Aggregated Attributions for Explanatory Analysis of 3D Segmentation Models Chrabaszcz, Maciej Baniecki, Hubert Komorowski, Piotr Płotka, Szymon Biecek, Przemyslaw Computer Vision and Pattern Recognition Analysis of 3D segmentation models, especially in the context of medical imaging, is often limited to segmentation performance metrics that overlook the crucial aspect of explainability and bias. Currently, effectively explaining these models with saliency maps is challenging due to the high dimensions of input images multiplied by the ever-growing number of segmented class labels. To this end, we introduce Agg^2Exp, a methodology for aggregating fine-grained voxel attributions of the segmentation model's predictions. Unlike classical explanation methods that primarily focus on the local feature attribution, Agg^2Exp enables a more comprehensive global view on the importance of predicted segments in 3D images. Our benchmarking experiments show that gradient-based voxel attributions are more faithful to the model's predictions than perturbation-based explanations. As a concrete use-case, we apply Agg^2Exp to discover knowledge acquired by the Swin UNEt TRansformer model trained on the TotalSegmentator v2 dataset for segmenting anatomical structures in computed tomography medical images. Agg^2Exp facilitates the explanatory analysis of large segmentation models beyond their predictive performance. The source code is publicly available at https://github.com/mi2datalab/agg2exp. |
| title | Aggregated Attributions for Explanatory Analysis of 3D Segmentation Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.16653 |