Learning a Clinically-Relevant Concept Bottleneck for Lesion Detection in Breast Ultrasound
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911938076737536 |
|---|---|
| author | Bunnell, Arianna Glaser, Yannik Valdez, Dustin Wolfgruber, Thomas Altamirano, Aleen González, Carol Zamora Hernandez, Brenda Y. Sadowski, Peter Shepherd, John A. |
| author_facet | Bunnell, Arianna Glaser, Yannik Valdez, Dustin Wolfgruber, Thomas Altamirano, Aleen González, Carol Zamora Hernandez, Brenda Y. Sadowski, Peter Shepherd, John A. |
| contents | Detecting and classifying lesions in breast ultrasound images is a promising application of artificial intelligence (AI) for reducing the burden of cancer in regions with limited access to mammography. Such AI systems are more likely to be useful in a clinical setting if their predictions can be explained to a radiologist. This work proposes an explainable AI model that provides interpretable predictions using a standard lexicon from the American College of Radiology's Breast Imaging and Reporting Data System (BI-RADS). The model is a deep neural network featuring a concept bottleneck layer in which known BI-RADS features are predicted before making a final cancer classification. This enables radiologists to easily review the predictions of the AI system and potentially fix errors in real time by modifying the concept predictions. In experiments, a model is developed on 8,854 images from 994 women with expert annotations and histological cancer labels. The model outperforms state-of-the-art lesion detection frameworks with 48.9 average precision on the held-out testing set, and for cancer classification, concept intervention is shown to increase performance from 0.876 to 0.885 area under the receiver operating characteristic curve. Training and evaluation code is available at https://github.com/hawaii-ai/bus-cbm. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_00267 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Learning a Clinically-Relevant Concept Bottleneck for Lesion Detection in Breast Ultrasound Bunnell, Arianna Glaser, Yannik Valdez, Dustin Wolfgruber, Thomas Altamirano, Aleen González, Carol Zamora Hernandez, Brenda Y. Sadowski, Peter Shepherd, John A. Computer Vision and Pattern Recognition Machine Learning Detecting and classifying lesions in breast ultrasound images is a promising application of artificial intelligence (AI) for reducing the burden of cancer in regions with limited access to mammography. Such AI systems are more likely to be useful in a clinical setting if their predictions can be explained to a radiologist. This work proposes an explainable AI model that provides interpretable predictions using a standard lexicon from the American College of Radiology's Breast Imaging and Reporting Data System (BI-RADS). The model is a deep neural network featuring a concept bottleneck layer in which known BI-RADS features are predicted before making a final cancer classification. This enables radiologists to easily review the predictions of the AI system and potentially fix errors in real time by modifying the concept predictions. In experiments, a model is developed on 8,854 images from 994 women with expert annotations and histological cancer labels. The model outperforms state-of-the-art lesion detection frameworks with 48.9 average precision on the held-out testing set, and for cancer classification, concept intervention is shown to increase performance from 0.876 to 0.885 area under the receiver operating characteristic curve. Training and evaluation code is available at https://github.com/hawaii-ai/bus-cbm. |
| title | Learning a Clinically-Relevant Concept Bottleneck for Lesion Detection in Breast Ultrasound |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2407.00267 |