Survey of explainable artificial intelligence techniques for biomedical imaging with deep neural networks

In the paper

“It was shown that LRP was useful for explainability of Alzheimer's classification predictions and applicable to similar disease diagnosis using MRI data [144].”

What the sources say

• Partly supported

  The Discussion and Conclusion show that LRP helped explain individual CNN decisions in MRI-based Alzheimer’s classification. The Conclusion says the framework *might* be useful for other disease-classification studies, but the paper does not demonstrate its applicability to diagnosing other diseases. (LLM, conf=0.85)

  cites [144] Böhle, 2019, Layer-wise relevance propagation for explaining deep neural network decisions in MRI-based Alzheimer's disease classification, Front. Aging Neurosci. sourceExperimental — accuracy not yet measured

In the paper

“The clinical acceptance of AI techniques would be aided by multi-modality and information fusion of both imaging and non-imaging data [34,170].”

What the sources say

• Not checked yetcites [34] Karthik, 2021, AI for COVID-19 detection from radiographs: incisive analysis of state of the art techniques, IRBM source
• Partly supported

  The Introduction and Discussion report that combining a chest-radiograph CNN score with a clinician’s assessment improved triage performance and could be useful in clinical practice. They do not assess multimodal imaging, fusion with non-imaging patient data, or whether either would increase clinical acceptance of AI. (LLM, conf=0.82)

  cites [170] Dunnmon, 2019, Assessment of convolutional neural networks for automated classification of chest radiographs, Radiology sourceExperimental — accuracy not yet measured

In the paper

“The increased use of XAI techniques is seen in neurology, respiratory and ophthalmic studies.”

In the paper

“CXR images are most favored for respiratory studies due to the ease with which these can be obtained, with Grad-CAM as the most prevalent XAI methods.”

In the paper

“In ophthalmic studies, fundus images seem to be the dominant modality.”

In the paper

“In neurology, the most prevalent image modality is MRI.”

In the paper

“Due to the paucity of the medical data, some research is focused on addressing the associated challenges [140,227].”

What the sources say

• Not checked yetcites [140] Choi, 2020, Cognitive signature of brain FDG PET based on deep learning: domain transfer from Alzheimer's disease to Parkinson's disease, Eur. J. Nucl. Med. Mol. Imag. source
• Not checked yetcites [227] Rajpurkar, 2020, AppendiXNet: deep learning for diagnosis of appendicitis from a small dataset of CT exams using video pretraining, Sci. Rep. source

In the paper

“The problem of bias can be addressed by an open sharing of the research process, methods, data and results, with such open sharing across institutions and funding agencies enabling open science [39].”

What the sources say

• Not checked yetcites [39] Norori, 2021, Addressing bias in big data and AI for health care: a call for open science, Patterns source

In the paper

“AI can potentially transform the image-based diagnosis in radiology [8].”

What the sources say

• Not checked yetcites [8] Folke, 2021, Explainable AI for medical imaging: explaining pneumothorax diagnoses with Bayesian teaching, Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications III

In the paper

“Neural networks require a lot of training data and time before they can produce acceptable results but are like a 'black-box' providing little or no insight into the decision-making process [13].”

What the sources say

• Not checked yetcites [13] McCarthy, 2021, Enterprise imaging and big data: a review from a medical physics perspective, Phys. Med. source

In the paper

“Convolutional Neural Networks (CNN) are state-of-the-art models for image segmentation and classification [2], although encoder-decoder transformer architectures have also been proposed [3][4][5].”

What the sources say

• Not checked yetcites [3] Ahamed, 2022, Joint learning with local and global consistency for improved medical image segmentation
• Not checked yetcites [4] Dosovitskiy, 2010
• Not checked yetcites [5] Strudel, 2021, Segmenter: transformer for semantic segmentation

In the paper

“The open models can be trained with local data from, for example, hospitals across the globe to mitigate bias and boost sample size [39].”

What the sources say

• Not checked yetcites [39] Norori, 2021, Addressing bias in big data and AI for health care: a call for open science, Patterns source

In the paper

“Algorithm or model bias can be a risk in AI as the models can generate a wrongful output based on an intentional or unintentional bias [40].”

What the sources say

• Not checked yetcites [40] Meske, 2020

In the paper

“Disease diagnosis through these techniques is inevitable to help overcome the shortage of, for example, 2000 radiologists in UK [9].”

What the sources say

• Not checked yetcites [9]

In the paper

“The performance of IBM Watson system for oncology had very successful performance and agreements with oncologists in some areas but poor in other areas, indicating adaptations to local practices and not extensible at a global scale [38].”

What the sources say

• Not checked yetcites [38] Durán, 2021

In the paper

“Clinicians are the primary users of XAI systems and they may have differing views but all have a need for the explainability for the patient [27].”

What the sources say

• Not checked yetcites [27] Antoniadi, 2021, Current challenges and future opportunities for XAI in machine learning-based clinical decision support systems: a systematic review, Appl. Sci. source

In the paper

“Most XAI work is by the AI experts which might help engineers to build better AI systems but may be limited for other stakeholders especially the end-user [8].”

What the sources say

• Not checked yetcites [8] Folke, 2021, Explainable AI for medical imaging: explaining pneumothorax diagnoses with Bayesian teaching, Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications III

In the paper

“The regulators would most likely be interested in an overall performance and a combination of explanations of the model rather than specific cases of individual patients [25].”

What the sources say

• Not checked yetcites [25] W. Samek and K. R. Müller, “Towards explainable artificial intelligence,” in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning. Lecture Notes in Computer Science, vol. vol. 11700, Springer, Cham..

In the paper

“The models with an intrinsic interpretability provide better explainability at lower predictive performance compared to post-hoc methods that in contrast have less explainability but a higher predictive performance [59].”

What the sources say

• Not checked yetcites [59] Bauer, 2021, Expl(AI)n it to me – explainable AI and information systems research, Business & Information Systems Engineering source

In the paper

“It was demonstrated that the proposed 2D capsule network system provided better visualizations and malignancy predictions compared to 3D CNN and achieved better diagnostic accuracy [15].”

What the sources say

• Not checked yetcites [15] R. LaLonde, D. Torigian and U. Bagci, “Encoding visual attributes in capsules for explainable medical diagnoses,” in Medical Image Computing and Computer Assisted Intervention – MICCAI 2020. source

In the paper

“For example, VGG-16 model requires 15.5 billion floating-point operations for a single image classification with approximately 138 million parameters [56].”

What the sources say

• Not checked yetcites [56] Yeom, 2021, Pruning by explaining: a novel criterion for deep neural network pruning, Pattern Recogn. source

In the paper

“Failure to identify and address the concerns of the system users can lead to another “AI Winter”, where a small market segment could change the research and development focus from XAI [61].”

What the sources say

• Not checked yetcites [61] A. Preece, D. Harborne, D. Braines, R. Tomsett and S. Chakraborty, “Stakeholders in Explainable AI,” [Online]. Available: https://doi.org/10.48550/arXiv.1810.00184. [Accessed 15 September 2022].

In the paper

“Natural Image Compression and Explanation (NICE) uses a mask over the entire image for each pixel in the image that captures the saliency based on its contribution to the prediction.”

What the sources say

• Not checked yetcites [75] Li, 2020, Neural image compression and explanation, IEEE Access source

In the paper

“While trustworthiness should be a property of every explainable system, this does not mean that every trustworthy system should also be explainable [2].”

What the sources say

• Not checked yetcites [2] Barredo Arrieta, 2020, Explainable Artificial Intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI, Inf. Fusion source

In the paper

“Self-explaining AI is proposed that produces a decision and an explanation.It provides a human understandable explanation and confidence level for each decision and which was deemed to motivate and lead towards trustworthy AI [55].”

What the sources say

• Not checked yetcites [55] Elton, 2020, Self-explaining AI as an alternative to interpretable AI

In the paper

“The heatmaps are commonly used to explain AI models and the quality of heatmaps can be appreciated using perturbation analysis [25].”

What the sources say

• Not checked yetcites [25] W. Samek and K. R. Müller, “Towards explainable artificial intelligence,” in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning. Lecture Notes in Computer Science, vol. vol. 11700, Springer, Cham..

In the paper

“The introduction of a hierarchy excluded the regions with less important features [74].”

What the sources say

• Not checked yetcites [74] Monroe, 2021, HihO: accelerating artificial intelligence interpretability for medical imaging in IoT applications using hierarchical occlusion, Neural Comput. Appl. source

In the paper

“Similar to ISO 13485:2016 Medical devices standard for Quality management systems that prescribes the medical devices regulatory requirements, there is a need for certifications of DNN models for clinical adoption [64].”

What the sources say

• Not checked yetcites [64] 2018

In the paper

“Local Interpretable Modelagnostic Explanations (LIME) was proposed to explain any classifier's predictions by approximating it locally around the prediction [70].”

What the sources say

• Not checked yetcites [70] Ribeiro, 2016, Why should i trust you?': explaining the predictions of any classifier

In the paper

“The patients require simplified explanations for the clinical decision as the DNN models are very complex and cannot be explained to a lay user [49].”

What the sources say

• Not checked yetcites [49] Gryz, 2021, Black box algorithms and the rights of individuals: no easy solution to the ‘explainability’ problem, Internet Policy Review source

In the paper

“Similarly, formal mathematical notations and techniques can unambiguously provide explanations for domain experts but are not targeted to the end users [63].”

What the sources say

• Not checked yetcites [63] Heuillet, 2021, Explainability in deep reinforcement learning, Knowl. Base Syst. source

In the paper

“The application developers would like to use the XAI techniques to understand the inner working and model improvement [62].”

What the sources say

• Not checked yetcites [62] Wang, 2020, CNN hyperparameter optimization based on CNN visualization and perception hash algorithm

In the paper

“Certification frameworks for a formal validation of models could testify the model operation statistically and ensure that the model predictions meet the required regulatory standards such as ISO/IEC CD 23053 (2020) for establishing a framework for use of AI systems [49].”

What the sources say

• Not checked yetcites [49] Gryz, 2021, Black box algorithms and the rights of individuals: no easy solution to the ‘explainability’ problem, Internet Policy Review source

In the paper

“Randomized Input Sampling for Explanation of Black-box Models (RISE) generates an importance map based on the importance of each pixel to the prediction [72].”

What the sources say

• Not checked yetcites [72] V. Petsiuk, A. Das and K. Saenko, “RISE: Randomized Input Sampling for Explanation of Black-box Models,” [Online]. Available: https://doi.org/10.48550/arXiv.1806.07421. [Accessed 15 September 2022].

In the paper

“A novel hierarchical occlusion algorithm is proposed and is based on two observations; that the important features are localized in an image, and features collection in multiple scales can affect the prediction outcome [74].”

What the sources say

• Not checked yetcites [74] Monroe, 2021, HihO: accelerating artificial intelligence interpretability for medical imaging in IoT applications using hierarchical occlusion, Neural Comput. Appl. source

In the paper

“Iterative and Adaptive Sampling with Spatial Attention (IASSA) was proposed that used an iterative and adaptive sampling module to determine the saliency map of each pixel for the model prediction [73].”

What the sources say

• Not checked yetcites [73] Vasu, 2020, Iterative and adaptive sampling with spatial attention for black-box model explanations

In the paper

“An interpretable model can obtain transparency by providing the reasons behind their output [52].”

What the sources say

• Not checked yetcites [52] Abeyrathna, 2021, Extending the tsetlin machine with integer-weighted clauses for increased interpretability, IEEE Access source

In the paper

“A novel hierarchical occlusion algorithm is proposed and is based on two observations; that the important features are localized in an image, and features collection in multiple scales can affect the prediction outcome [74].”

What the sources say

• Not checked yetcites [74] Monroe, 2021, HihO: accelerating artificial intelligence interpretability for medical imaging in IoT applications using hierarchical occlusion, Neural Comput. Appl. source

In the paper

“There is little agreement on what is an interpretable ML system and how it should be evaluated [51].”

What the sources say

• Not checked yetcites [51] Doshi-Velez, 2017

In the paper

“There is a need to have a more formalized definition of interpretability in the context of DNNs [51].”

What the sources say

• Not checked yetcites [51] Doshi-Velez, 2017

In the paper

“A lack of consensus in the meaning of explainability originates due to different stakeholders, each having their own understanding of the terms [61].”

What the sources say

• Not checked yetcites [61] A. Preece, D. Harborne, D. Braines, R. Tomsett and S. Chakraborty, “Stakeholders in Explainable AI,” [Online]. Available: https://doi.org/10.48550/arXiv.1810.00184. [Accessed 15 September 2022].

In the paper

“There is a trade-off between the accuracy and interpretability of the models, as shown in Fig. 2, where non-linear models with better performance, such as DNNs are less explainable, and linear models which are least accurate, such as decision trees, are more explainable [2,21,45].”

What the sources say

• Not checked yetcites [2] Barredo Arrieta, 2020, Explainable Artificial Intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI, Inf. Fusion source
• Not checked yetcites [21] Adadi, 2018, Peeking inside the black-box: a survey on explainable artificial intelligence (XAI), IEEE Access source
• Not checked yetcites [45] Gunning, 2019

In the paper

“The model during training ideally should learn the discriminating image features of the various classes but instead can also learn extraneous features [42].”

What the sources say

• Not checked yetcites [42] Zech, 2018, Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: a cross-sectional study, PLoS Med. source

In the paper

“For domains with a lack of knowledge about the underlying causal process, such as healthcare, it is prudent not to prefer explainability over diagnostic or predictive accuracy as this may affect the health and well-being of patients with some who get additional investigations and treatment and similarly those who are mjs-or undiagnosed [57].”

What the sources say

• Not checked yetcites [57] London, 2019, Artificial intelligence and black-box medical decisions: accuracy versus explainability, Hastings Cent. Rep. source

In the paper

“The proposed technique exploited learned attention for better interpretability, well-aligned to human intuition on the datasets with METEOR and BLEU metrics [92].”

What the sources say

• Not checked yetcites [92] Xu, 2015

In the paper

“The LRP has some limitations such as the heatmaps were noisy and could be non-discriminative [84].”

What the sources say

• Not checked yetcites [84] Jung, 2021, Explaining CNN and RNN using selective layer-wise relevance propagation, IEEE Access source

In the paper

“Among the evaluated XAI methods on the evaluation task, Deconvnet and Grad-CAM were found to be least accurate, whereas LRP, Integrated Gradients and Guided Backpropagation were found to be the most accurate methods [130].”

What the sources say

• Not checked yetcites [130] Arras, 2022, CLEVR-XAI: a benchmark dataset for the ground truth evaluation of neural network explanations, Inf. Fusion source

In the paper

“The evaluation of proposed technique was found to be more consistent with human understanding and intuition with an improved computational efficiency compared to the other selected approaches [90].”

What the sources say

• Not checked yetcites [90] Lundberg, 2017, A unified approach to interpreting model predictions, Adv. Neural Inf. Process. Syst.

In the paper

“Gradient-weighted Class Activation Mapping (Grad-CAM) was proposed for producing visual explanations by using the gradients in the final model layer to identify and highlight the important image regions [79].”

What the sources say

• Not checked yetcites [79] Selvaraju, 2017, Visual explanations from deep networks via gradient-based localization

In the paper

“It was demonstrated that the proposed method could work in the presence of adversarial noise and irrespective of the model accuracy [82].”

What the sources say

• Not checked yetcites [82] Muhammad, 2020, Eigen-CAM: class activation map using principal components

In the paper

“The method was motivated by Grad-CAM and RISE, however, unlike Grad-CAM it was gradient-free, and unlike RISE, it used a combined final mask instead of a random mask [76].”

What the sources say

• Not checked yetcites [76] S. M. Muddamsetty, N. S. J. Mohammad and T. B. Moeslund, “SIDU: similarity difference and uniqueness method for explainable AI,” in IEEE International Conference on Image Processing (ICIP).

In the paper

“The model was evaluated through a comparison with LRP, Contrastive LRP (CLRP), and Softmax Gradient LRP (SGLRP), and found to provide less noise, better class discrimination and held entire target objects [84].”

What the sources say

• Not checked yetcites [84] Jung, 2021, Explaining CNN and RNN using selective layer-wise relevance propagation, IEEE Access source

In the paper

“It was shown that the combined approach achieved better attention maps and a higher classification accuracy [114].”

What the sources say

• Not checked yetcites [114] Kashyap, 2020, Looking in the right place for anomalies: explainable ai through automatic location learning

In the paper

“The interface was evaluated using surveys for both experts of DNN and non-experts and showed the approach to be effective [126].”

What the sources say

• Not checked yetcites [126] Vyas, 2020, An interactive graphical visualization approach to CNNs and RNNs

In the paper

“It was argued that the Gradient-Based methods are lacking as the neurons' gradients do not relate to the top-down saliency [85].”

What the sources say

• Not checked yetcites [85] Zhang, 2018, Top-down neural attention by excitation Backprop, Int. J. Comput. Vis. source

In the paper

“Grad-CAM++ was proposed as a method to build on Grad-CAM and provided better visual explanations compared to it and was better in explaining multiple occurrences of the same class objects in a single image [80].”

What the sources say

• Not checked yetcites [80] Chattopadhay, 2018, Grad-CAM++: generalized gradient-based visual explanations for deep convolutional networks

In the paper

“The best performing individual DNN architectures were VGG-19 and DenseNet-161 and the heatmaps generated by Grad-CAM++ were the most reliable in comparison with Grad-CAM and LRP [162].”

What the sources say

• Not checked yetcites [162] Karim, 2020, Deepcovidexplainer: explainable COVID-19 diagnosis from chest X-ray images

In the paper

“A Hierarchical Feature Network (HFnet) was proposed that showed that the low-level layer features are superior in comparison to low-level semantic questions [115].”

What the sources say

• Not checked yetcites [115] Hong, 2019, Exploiting hierarchical visual features for visual question answering, Neurocomputing source

In the paper

“It was found that any foreign substance in the lung cavity resulted in misclassification of images, and sometimes a correct classification was due to the wrong reasons [164].”

What the sources say

• Not checked yetcites [164] Lee, 2020, Evaluation of scalability and degree of fine-tuning of deep convolutional neural networks for COVID-19 screening on chest X-ray images using explainable deep-learning algorithm, J. Personalized Med. source

In the paper

“The model performance was more accurate on CT compared to CXR images, and had an accuracy of 95.61% in COVID-19 detection.”

What the sources say

• Not checked yetcites [158] Panwar, 2020

In the paper

“Thus, the gradient based methods lack in generalizability and robustness [85].”

What the sources say

• Not checked yetcites [85] Zhang, 2018, Top-down neural attention by excitation Backprop, Int. J. Comput. Vis. source

In the paper

“Compared to qualitative approaches through visual observations and comparisons, quantitative approaches can provide more objectivity and reliability in the process.”

In the paper

“The framework was qualitatively evaluated on MNIST dataset with different types of human users and received very positive feedback [124].”

What the sources say

• Not checked yetcites [124] Spinner, 2020, IEEE Trans. Visual. Comput. Graph.

In the paper

“It was shown through visualizations that the features are not random or uninterpretable patterns but rather have class discrimination and increasing invariance between layers [87].”

What the sources say

• Not checked yetcites [87] Zeiler, 2014, Visualizing and understanding convolutional networks

In the paper

“A visual explainable system Similarity Difference and Uniqueness Method (SIDU) was proposed to estimate pixel saliency by extracting feature maps from the last CNN layer, and creating a difference mask to form the visual explanation of the prediction [76].”

What the sources say

• Not checked yetcites [76] S. M. Muddamsetty, N. S. J. Mohammad and T. B. Moeslund, “SIDU: similarity difference and uniqueness method for explainable AI,” in IEEE International Conference on Image Processing (ICIP).

In the paper

“The proposed scheme was evaluated on clinical and general datasets and showed better visual explanations with a greater trust for the human expert [76].”

What the sources say

• Not checked yetcites [76] S. M. Muddamsetty, N. S. J. Mohammad and T. B. Moeslund, “SIDU: similarity difference and uniqueness method for explainable AI,” in IEEE International Conference on Image Processing (ICIP).

In the paper

“COVID-19 detection on CXR images with a fine tuned EfficientNet-B7 achieved an accuracy of 95.01% and AUC scores of 0.950, which were better in comparison to many other DNN approaches for COVID-19 detection [161].”

What the sources say

• Not checked yetcites [161] Chetoui, 2021, Deep efficient neural networks for explainable COVID-19 detection on CXR images

In the paper

“Despite the large number of existing explainable models, there is no agreement on techniques for measuring or evaluating the explainability performance of an AI model.”

In the paper

“The two explainability approaches are complementary, in that, saliency map highlights the pixels with significant contribution to the model prediction and explanation-by-examples selects the training images likely for prediction in Bayesian Teaching [8].”

What the sources say

• Not checked yetcites [8] Folke, 2021, Explainable AI for medical imaging: explaining pneumothorax diagnoses with Bayesian teaching, Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications III

In the paper

“An adapted ResNet architecture was used on MR CLEAN Registry dataset of CT angiography images and determined that the automated DNN method was better than radiological image biomarkers for stroke outcome prediction and was helpful in the treatment selection [136].”

What the sources say

• Not checked yetcites [136] Hilbert, 2019, Data-efficient deep learning of radiological image data for outcome prediction after endovascular treatment of patients with acute ischemic stroke, Comput. Biol. Med. source

In the paper

“The feature map showed positive values for the portions with significant influence on the disease diagnosis [236].”

What the sources say

• Not checked yetcites [236] Yoo, 2021, Diagnosis support model of cardiomegaly based on CNN using ResNet and explainable feature map, IEEE Access source

In the paper

“The CT images from multiple sources were mixed to avoid overfitting and bias [175].”

What the sources say

• Not checked yetcites [175] Wu, 2021, JCS: an explainable COVID-19 diagnosis system by Joint classification and segmentation, IEEE Trans. Image Process. source

In the paper

“The XAI can help in identifying all such factors which form the basis of a particular decision.”

In the paper

“The method provided insights of the high-level feature importance in the trained CNN model [224].”

What the sources say

• Not checked yetcites [224] Couteaux, 2019, Towards interpretability of segmentation networks by analyzing deepdreams

In the paper

“The proposed architecture provided better insights into the model's decisions and achieved better performance compared to the considered approaches.”

What the sources say

• Not checked yetcites [182] Kumar, 2019, SISC: end-to-end interpretable discovery radiomics-driven lung cancer prediction via stacked interpretable sequencing cells, IEEE Access source

In the paper

“The ensemble was found to improve the classification of the system and achieved an AUC of 90.29% in predicting nodules that would probably be diagnosed as cancerous in two years.”

What the sources say

• Not checked yetcites [179] Paul, 2020, Convolutional Neural Network ensembles for accurate lung nodule malignancy prediction 2 years in the future, Comput. Biol. Med. source

In the paper

“However, the dataset was small with low-resolution images [167].”

What the sources say

• Not checked yetcites [167] Liz, 2021, Ensembles of Convolutional Neural Network models for pediatric pneumonia diagnosis, Future Generat. Comput. Syst. source

In the paper

“Emergent Symbolic language are the cornerstones of Good Old-Fashioned AI (GOFAI) systems which are explainable [252].”

What the sources say

• Not checked yetcites [252] Chowdhury, 2021, Emergent Symbolic language based deep medical image classification

In the paper

“CXR images were used for lung cancer prediction using the proposed X-Caps XAI technique generating malignancy scores with explanations at human-level that were better interpretable by the radiologists [15].”

What the sources say

• Not checked yetcites [15] R. LaLonde, D. Torigian and U. Bagci, “Encoding visual attributes in capsules for explainable medical diagnoses,” in Medical Image Computing and Computer Assisted Intervention – MICCAI 2020. source

In the paper

“The JSC identified and segmented the opacification areas, a basic CT feature of COVID-19 [175].”

What the sources say

• Not checked yetcites [175] Wu, 2021, JCS: an explainable COVID-19 diagnosis system by Joint classification and segmentation, IEEE Trans. Image Process. source

In the paper

“Emergent languages involve multi-agent communication games with a sender/ receiver coordinate to complete a task and have been shown to enhance the neural network capabilities [252].”

What the sources say

• Not checked yetcites [252] Chowdhury, 2021, Emergent Symbolic language based deep medical image classification

In the paper

“The model performance was found comparable to similar 3D CNN models and had a high diagnostic performance for acute appendicitis detection [227].”

What the sources say

• Not checked yetcites [227] Rajpurkar, 2020, AppendiXNet: deep learning for diagnosis of appendicitis from a small dataset of CT exams using video pretraining, Sci. Rep. source

In the paper

“The proposed radiomic sequencer achieved correct predictions that can potentially improve clinical adoption [182].”

What the sources say

• Not checked yetcites [182] Kumar, 2019, SISC: end-to-end interpretable discovery radiomics-driven lung cancer prediction via stacked interpretable sequencing cells, IEEE Access source

In the paper

“The proposed system achieved an AUC of 0.9627 and a detection accuracy of 0.9272 [240].”

What the sources say

• Not checked yetcites [240] Huo, 2019, Coronary calcium detection using 3D attention identical dual deep network based on weakly supervised learning, Med. Imaging Image Process.

In the paper

“This can also help in overcoming the challenges associated with the limited information present in an X-ray and actually highlighting how the density, shape, size and location of abnormalities affect the final diagnosis.”

In the paper

“RFGB found simpler rules for modelling the features and could be easily compared to a large single highly accurate model [219].”

What the sources say

• Not checked yetcites [219] Kunapuli, 2018, A decision-support tool for renal mass classification,” journal of digital imaging, J. Digit. Imag. source

In the paper

“An average classification accuracy of 94.78% was achieved.”

What the sources say

• Not checked yetcites [177] Xu, 2019, Pulmonary textures classification via a multi-scale attention network, IEEE journal of biomedical and health informatics source

In the paper

“It was shown that the hierarchical diagnosis performed better than the other models for classification, and identified the relevant image regions with attention maps [217].”

What the sources say

• Not checked yetcites [217] Barata, 2021, Explainable skin lesion diagnosis using taxonomies, Pattern Recogn. source

In the paper

“The ABCD-rule of dermoscopy is known to outperform other methods [212].”

What the sources say

• Not checked yetcites [212] Stieler, 2021, Towards domain-specific explainable AI:model interpretation of a skin image classifier using a human approach

In the paper

“The study concluded that by embedding model explainability in the early model training can save time and help clinicians in using a model's prediction for the clinical decisions [152].”

What the sources say

• Not checked yetcites [152] Windisch, 2020, Implementation of model explainability for a basic brain tumor detection using convolutional neural networks on MRI slices, Neuroradiology source

In the paper

“Four CNN models were used and visualizations by heatmaps determined the important features which were compared against the expert's judgement [187].”

What the sources say

• Not checked yetcites [187] Reguant, 2021, Understanding inherent image features in CNN-based assessment of diabetic retinopathy, Sci. Rep. source

In the paper

“From the aggregated explanations it was concluded that cisterns and ventricles governed the age predictions [146].”

What the sources say

• Not checked yetcites [146] Levakov, 2020, From a deep learning model back to the brain—identifying regional predictors and their relation to aging, Hum. Brain Mapp. source

In the paper

“It can be difficult to control the image quality due to manual operation, operator dependency and acoustic shadows [253].”

What the sources say

• Not checked yetcites [253] Komatsu, 2021, Towards clinical application of artificial intelligence in ultrasound imaging, Biomedicines source

In the paper

“CA-Net also provided better explainability by visualizing the attention weight maps, although it had around 15 times smaller model size [153].”

What the sources say

• Not checked yetcites [153] Lopatina, 2020, Investigation of deep-learning-driven identification of multiple sclerosis patients based on susceptibility-weighted images using relevance analysis, Front. Neurosci. source

In the paper

“The significant features for the prediction were identified by isolating the neuron activation patterns [192].”

What the sources say

• Not checked yetcites [192] Niu, 2021, Explainable diabetic retinopathy detection and retinal image generation, IEEE journal of biomedical and health informatics

In the paper

“Grad-CAM was used for visualisations and helped to identify the tumor location.”

What the sources say

• Not checked yetcites [152] Windisch, 2020, Implementation of model explainability for a basic brain tumor detection using convolutional neural networks on MRI slices, Neuroradiology source

In the paper

“The video data was converted to a barcode type timeline to aid the explainability which made it easier to identify the substructures [243].”

What the sources say

• Not checked yetcites [243] Komatsu, 2021, Detection of cardiac structural abnormalities in fetal ultrasound videos using deep learning, Appl. Sci. source

In the paper

“The datasets EyePACS and DIARETDB1 were used and although different CNN models had a consistent classification performance, the image feature disagreement between different architectures was 70% as the highest value [187].”

What the sources say

• Not checked yetcites [187] Reguant, 2021, Understanding inherent image features in CNN-based assessment of diabetic retinopathy, Sci. Rep. source

In the paper

“The choice of the MRI slice/image is important and the authors proposed selecting the best MRI slice for improved model performance [220].”

What the sources say

• Not checked yetcites [220] Hassan, 2022, Prostate cancer classification from ultrasound and MRI images using deep learning based Explainable Artificial Intelligence, Future Generat. Comput. Syst. source

In the paper

“Breast ultrasound and MRI can provide additional information for the mammographic diagnosis [254].”

What the sources say

• Not checked yetcites [254] Le, 2019, Artificial intelligence in breast imaging, Clin. Radiol. source