Convolutional attention network for MRI-based Alzheimer's disease classification and its interpretability analysis
dc.authorid | 0000-0002-8649-6013 | |
dc.authorid | 0000-0002-8649-6013 | en_US |
dc.contributor.author | Türkan, Yasemin | en_US |
dc.contributor.author | Tek, Faik Boray | en_US |
dc.date.accessioned | 2022-05-23T17:00:46Z | |
dc.date.available | 2022-05-23T17:00:46Z | |
dc.date.issued | 2021-09-17 | |
dc.department | Işık Üniversitesi, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümü | en_US |
dc.department | Işık University, Faculty of Engineering, Department of Computer Engineering | en_US |
dc.description.abstract | Neuroimaging techniques, such as Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET), help to identify Alzheimer's disease (AD). These techniques generate large-scale, high-dimensional, multimodal neuroimaging data, which is time-consuming and difficult to interpret and classify. Therefore, interest in deep learning approaches for the classification of 3D structural MRI brain scans has grown rapidly. In this research study, we improved the 3D VGG model proposed by Korolev et al. [2]. We increased the filters in the 3D convolutional layers and then added an attention mechanism for better classification. We compared the performance of the proposed approaches for the classification of Alzheimer's disease versus mild cognitive impairments and normal cohorts on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. We observed that both the accuracy and area under curve results improved with the proposed models. However, deep neural networks are black boxes that produce predictions that require further explanation for medical usage. We compared the 3D-data interpretation capabilities of the proposed models using four different interpretability methods: Occlusion, 3D Ultrametric Contour Map, 3D Gradient-Weighted Class Activation Mapping, and SHapley Additive explanations (SHAP). We observed that explanation results differed in different network models and data classes. | en_US |
dc.identifier.citation | Türkan, Y. & Tek, F. B. (2021). Convolutional attention network for MRI-based Alzheimer's disease classification and its interpretability analysis. Paper presented at the 2021 6th International Conference on Computer Science and Engineering (UBMK), 151-156. doi:10.1109/UBMK52708.2021.9558882 | en_US |
dc.identifier.doi | 10.1109/UBMK52708.2021.9558882 | |
dc.identifier.endpage | 156 | |
dc.identifier.isbn | 9781665429085 | |
dc.identifier.isbn | 9781665429078 | |
dc.identifier.isbn | 9781665429092 | |
dc.identifier.issn | 2521-1641 | en_US |
dc.identifier.issn | 2768-0592 | en_US |
dc.identifier.scopus | 2-s2.0-85125875116 | en_US |
dc.identifier.scopusquality | N/A | en_US |
dc.identifier.startpage | 151 | |
dc.identifier.uri | https://hdl.handle.net/11729/4351 | |
dc.identifier.uri | http://dx.doi.org/10.1109/UBMK52708.2021.9558882 | |
dc.indekslendigikaynak | Scopus | en_US |
dc.institutionauthor | Türkan, Yasemin | en_US |
dc.institutionauthor | Tek, Faik Boray | en_US |
dc.language.iso | en | en_US |
dc.peerreviewed | Yes | en_US |
dc.publicationstatus | Published | en_US |
dc.publisher | IEEE | en_US |
dc.relation.ispartof | 2021 6th International Conference on Computer Science and Engineering (UBMK) | en_US |
dc.relation.publicationcategory | Konferans Öğesi - Uluslararası - Kurum Öğretim Elemanı | en_US |
dc.rights | info:eu-repo/semantics/closedAccess | en_US |
dc.subject | 3D Modeling | en_US |
dc.subject | 3D Structural MRI brain scans | en_US |
dc.subject | 3D Ultrametric contour map | en_US |
dc.subject | 3D VGG model | en_US |
dc.subject | 3D-Data interpretation capabilities | en_US |
dc.subject | Activation analysis | en_US |
dc.subject | Activation mapping | en_US |
dc.subject | Alzheimer’s disease | en_US |
dc.subject | Alzheimers disease | en_US |
dc.subject | Attention | en_US |
dc.subject | Attention mechanism | en_US |
dc.subject | Biomedical MRI | en_US |
dc.subject | Brain | en_US |
dc.subject | Chemical activation | en_US |
dc.subject | Classification (of information) | en_US |
dc.subject | Cognition | en_US |
dc.subject | Contour map | en_US |
dc.subject | Convolution | en_US |
dc.subject | Convolutional attention network | en_US |
dc.subject | Convolutional networks | en_US |
dc.subject | Convolutional neural networks | en_US |
dc.subject | Deep learning | en_US |
dc.subject | Deep learning approaches | en_US |
dc.subject | Deep neural networks | en_US |
dc.subject | Different interpretability methods | en_US |
dc.subject | Different network models | en_US |
dc.subject | Diseases | en_US |
dc.subject | Heating systems | en_US |
dc.subject | High-dimensional neuroimaging data | en_US |
dc.subject | Image classification | en_US |
dc.subject | Interpretability | en_US |
dc.subject | Interpretability analysis | en_US |
dc.subject | Learning (artificial intelligence) | en_US |
dc.subject | Magnetic resonance imaging | en_US |
dc.subject | Mapping | en_US |
dc.subject | Medical image processing | en_US |
dc.subject | Mild cognitive impairments | en_US |
dc.subject | MRI | en_US |
dc.subject | MRI-based Alzheimer | en_US |
dc.subject | Multimodal neuroimaging data | en_US |
dc.subject | Neural nets | en_US |
dc.subject | Neurodegenerative diseases | en_US |
dc.subject | Neuroimaging | en_US |
dc.subject | Neuroimaging techniques | en_US |
dc.subject | Neurophysiology | en_US |
dc.subject | Occlusion | en_US |
dc.subject | Positron emission tomography | en_US |
dc.subject | SHAP | en_US |
dc.subject | Shapley | en_US |
dc.subject | Shapley additive explanation | en_US |
dc.subject | Solid modeling | en_US |
dc.subject | Three-dimensional displays | en_US |
dc.subject | Ultrametrics | en_US |
dc.subject | Visualization | en_US |
dc.title | Convolutional attention network for MRI-based Alzheimer's disease classification and its interpretability analysis | en_US |
dc.type | Conference Object | en_US |