Explainable CNN-attention Networks (C-Attention Network) for Automated Detection of Alzheimer's Disease
Wang, N.; CHEN, M.; Subbalakshmi, K. P.
Show abstract
AO_SCPLOWBSTRACTC_SCPLOWIn this work we propose three explainable deep learning architectures to automatically detect patients with Alzheimers disease based on their language abilities. The architectures use: (1) only the part-of-speech features; (2) only language embedding features and (3) both of these feature classes via a unified architecture. We use self-attention mechanisms and interpretable 1-dimensional Convolutional Neural Network (CNN) to generate two types of explanations of the models action: intra-class explanation and inter-class explanation. The inter-class explanation captures the relative importance of each of the different features in that class, while the inter-class explanation captures the relative importance between the classes. Note that although we have considered two classes of features in this paper, the architecture is easily expandable to more classes because of its modularity. Extensive experimentations and comparison with several recent models show that our method outperforms these methods with an accuracy of 92.2% and F1 score of 0.952 on the DementiaBank dataset while being able to generate explanations. We show by examples, how to generate these explanations using attention values.
Matching journals
The top 10 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Network-based identification of genetic factors in Ageing, lifestyle and Type 2 Diabetes that Influence in the progression of Alzheimer’s disease 89%
- A Combined Deep CNN-LSTM Network for the Detection of Novel Coronavirus (COVID-19) Using X-ray Images 89%
- DermoExpert: Skin lesion classification using a hybrid convolutional neural network through segmentation, transfer learning, and augmentation 89%
Similar papers in this journal
- Subtle anomaly detection in MRI brain scans: Application to biomarkers extraction in patients with de novo Parkinson’s disease 90%
- Deep ensemble multitask classification of emergency medical call incidents combining multimodal data improves emergency medical dispatch 90%
- Uncertainty in Deep Learning for EEG under Dataset Shifts 89%
Similar papers in this journal
- Practical Strategies for Extreme Missing Data Imputation in Dementia Diagnosis 91%
- pathCLIP: Detection of Genes and Gene Relations from Biological Pathway Figures through Image-Text Contrastive Learning 91%
- Evaluating Explanations from AI Algorithms for Clinical Decision-Making: A Social Science-based Approach 90%
Similar papers in this journal
- c-Triadem: A constrained, explainable deep learning model to identify novel biomarkers in Alzheimer’s disease 93%
- Interpretable multivariate survival models: Improving predictions for conversion from mild cognitive impairment to Alzheimers disease (AD) via data fusion and machine learning 93%
- Random forest model for feature-based Alzheimer's disease conversion prediction from early mild cognitive impairment subjects 93%
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.