Multi-modal volumetric concept activation to explain detection and classification of metastatic prostate cancer on PSMA-PET/CT
Not on the map. Why?
Abstract (arXiv)
Explainable artificial intelligence (XAI) is increasingly used to analyze the behavior of neural networks. Concept activation uses human-interpretable concepts to explain neural network behavior. This study aimed at assessing the feasibility of regression concept activation to explain detection and classification of multi-modal volumetric data. Proof-of-concept was demonstrated in metastatic prostate cancer patients imaged with positron emission tomography/computed tomography (PET/CT). Multi-modal volumetric concept activation was used to provide global and local explanations. Sensitivity was 80% at 1.78 false positive per patient. Global explanations showed that detection focused on CT for anatomical location and on PET for its confidence in the detection. Local explanations showed promise to aid in distinguishing true positives from false positives. Hence, this study demonstrated feasibility to explain detection and classification of multi-modal volumetric data using regression concept activation.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 10 of its 22 references that are in the map. First posted on arXiv 4 Aug 2022.
Builds on
10 of 22 references.
References · 10
- 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
- 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,400
- 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
- 2018Sanity Checks for Saliency Mapscited by 1,149
- 2019Interpretable machine learning: definitions, methods, and applicationscited by 303
- 2021Explainable artificial intelligence (XAI) in deep learning-based medical image analysiscited by 147
- 2021nnDetection: A Self-configuring Method for Medical Object Detectioncited by 60
- 2020Assessing the (Un)Trustworthiness of Saliency Maps for Localizing Abnormalities in Medical Imagingcited by 32
- 2019Testing the robustness of attribution methods for convolutional neural networks in MRI-based Alzheimer's disease classificationcited by 25
- 2021Leveraging Clinical Characteristics for Improved Deep Learning-Based Kidney Tumor Segmentation on CTcited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.