Explainable Deep Learning Methods in Medical Imaging Diagnosis: A Survey

Patrício, Cristiano; Neves, João C.; Teixeira, Luís

doi:10.48550/arxiv.2205.04766

Cited by 8 publications

(8 citation statements)

References 85 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For comparatively difficult medical image segmentation, five types of interpretation methods are especially summarized, including attribution map, concept attribution, language description, internal network representation and latent space interpretation. More recently, based on explanation modalities Patricio et al [18] categorized interpretation methods for DLB-MIA into five types: explanation by feature attribution, explanation by text, explanation by examples, explanation by concepts, and other approaches. The corresponding application paradigms are clearly illustrated by figures.…”

Section: Methods Taxonomiesmentioning

confidence: 99%

Towards reliable medical image analysis based on deep learning with XAI

Jiang

2023

International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023)

View full text Add to dashboard Cite

Deep learning-based medical image analysis (DLB-MIA) has achieved great success hitherto, but is in urgent need of model interpretability. Explainable artificial intelligence (XAI) can be used for improving model interpretability and is expected to help in providing crucial trustworthiness in DLB-MIA. To expedite the utilization of XAI technology for reliable DLB-MIA, the current status of XAI methods and toolkits is introduced firstly. The application of XAI technology in DLB-MIA tasks is then discussed with close attention in respect of method taxonomies and adversarial robustness respectively.

show abstract

Section: Methods Taxonomiesmentioning

confidence: 99%

Towards reliable medical image analysis based on deep learning with XAI

Jiang

2023

International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023)

View full text Add to dashboard Cite

show abstract

“…Medical practitioners can contribute vital insights into the clinical relevance and significance of specific features or patterns found in US images through collaborative efforts, allowing AI algorithms to generate more informed predictions. Incorporating real-time feedback and iterative learning into the diagnosis process can also improve diagnostic accuracy [140]. When physicians interact with the AI system and provide feedback on its predictions, the system can adapt and learn, refining its algorithms and improving its performance over time.…”

Section: Enhancing Diagnostic Accuracymentioning

confidence: 99%

Ultrasound imaging based recognition of prenatal anomalies: a systematic clinical engineering review

Sriraam,

Chinta,

Suresh

et al. 2024

Prog. Biomed. Eng.

View full text Add to dashboard Cite

For prenatal screening, ultrasound imaging allows for real-time observation of developing fetal anatomy. Understanding normal and aberrant forms through extensive fetal structural assessment enables for early detection and intervention. However, the reliability of anomaly diagnosis varies depending on operator expertise and device limits. First trimester scans in conjunction with circulating biochemical markers are critical in identifying high-risk pregnancies, but they also pose technical challenges. Recent engineering advancements in automated diagnosis, such as AI-based ultrasound image processing and multimodal data fusion, are developing to improve screening efficiency, accuracy, and consistency. Still, creating trust in these data-driven solutions is necessary for integration and acceptability in clinical settings. Transparency can be promoted by explainable AI (XAI) technologies that provide visual interpretations and illustrate the underlying diagnostic decision making process. An explanatory framework based on deeplearning is suggested to construct charts depicting anomaly screening results from ultrasound video feeds. AI modeling can then be applied to these charts to connect defects with probable deformations. Overall, engineering approaches that increaseimaging, automation, and interpretability hold enormous promise for altering traditional workflows and expanding diagnostic capabilities for better prenatal care.

show abstract

“…Saliency methods produce a visual interpretation map that represent the importance of image pixels for network classification. Class activation mapping is a pioneering saliency method [17]. [32] uses Global Average Pooling (GAP) to integrate information from all features to obtain CAM.…”

Section: Saliency Mapsmentioning

confidence: 99%

DProtoNet: Decoupling Prototype Activation via Multiple Dynamic Masks

Peng¹,

Liu²,

Yang³

2022

Preprint

View full text Add to dashboard Cite

The interpretability of neural networks has recently received extensive attention. The previous prototype-based explainable networks involved prototype activation in both the reasoning process and interpretation process, which requires specific explainable structures for the prototype. This makes the network less accurate as it gains interpretability. To avoid this problem, we propose a new model: decoupling prototypical network (DProtoNet), which contains three modules. 1) encoder module: we propose unrestricted masks to generate expressive features and prototypes. 2) inference module: we propose a multi-image prototype learning method to update prototypes so that the network can learn generalized prototypes.3) interpretation module: we propose multiple dynamic masks (MDM) decoder to explain the network, which generates heatmaps using the consistent activation of the original image and mask image at the detection nodes of the network. It decouples the inference module and interpretation module of a prototype-based network by avoiding the use of prototype activation to explain the network's decisions, so that the accuracy and interpretability of the network can be simultaneously improved. We test on multiple public general and medical datasets. The accuracy of our method is improved compared with the previous methods, which can be improved by up to 5%. DProtoNet achieves state-of-the-art interpretability.

show abstract

Explainable Deep Learning Methods in Medical Imaging Diagnosis: A Survey

Cited by 8 publications

References 85 publications

Towards reliable medical image analysis based on deep learning with XAI

Towards reliable medical image analysis based on deep learning with XAI

Ultrasound imaging based recognition of prenatal anomalies: a systematic clinical engineering review

DProtoNet: Decoupling Prototype Activation via Multiple Dynamic Masks

Contact Info

Product

Resources

About