{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,27]],"date-time":"2026-10-27T14:17:24Z","timestamp":1793110644604,"version":"4.5.0"},"reference-count":50,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T00:00:00Z","timestamp":1764028800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurosci."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Deep neural network architectures have transformed medical imaging, particularly in structural MRI (sMRI) classification. However, existing state-of-the-art deep learning models face limitations in preprocessing and feature extraction when classifying dementia-related conditions. This study addresses these challenges by evaluating multiple architectures for dementia diagnosis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study assessed eight pretrained convolutional neural networks (CNNs), a Vision Transformer (ViT), a multimodal attention model, and a capsule network (CapsNet) for classifying three classes: dementia, mild cognitive impairment (MCI), and healthy controls. The dataset, obtained from ADNI, was balanced across classes and comprised 10,000 training images per class, 3,000 validation images per class, and 850 test images per class. Classification was performed using 2D slices from sMRI scans. Performance metrics included accuracy, specificity, and sensitivity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Among all evaluated models, the 3D-CNN and multimodal attention models achieved the highest performance, with accuracies of 84% and 86%, specificities of 83% and 86%, and sensitivities of 84% and 86%, respectively. The ViT and CapsNet models achieved 100% sensitivity for Alzheimer\u2019s disease (AD) but demonstrated low precision for AD (43%) and 0% for other classes, indicating class imbalance effects. All models showed reduced performance and bias toward certain classes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The findings highlight the limitations of current architectures in sMRI dementia classification, including suboptimal feature extraction and class-specific biases. While certain models, such as multimodal attention and 3D-CNN, performed better overall, precision and generalization remain challenges. Future work should focus on improved data representation through advanced computer vision methods and architectural modifications to enhance diagnostic accuracy and computational efficiency.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnins.2025.1638022","type":"journal-article","created":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T14:46:06Z","timestamp":1764081966000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":7,"title":["Multi-model deep learning for dementia detection: addressing data and model limitations"],"prefix":"10.3389","volume":"19","author":[{"given":"Areej Y.","family":"Bayahya","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fares","family":"Jammal","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haneen","family":"Banjar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fathy Elbouraey","family":"Eassa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Omar","family":"Talabay","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sultan H.","family":"Alamri","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,11,25]]},"reference":[{"key":"ref1","doi-asserted-by":"crossref","DOI":"10.1109\/ICASSP.2019.8683759","article-title":"Capsule networks for brain tumor classification based on MRI images and coarse tumor boundaries","volume-title":"ICASSP 2019\u20132019 IEEE international conference on acoustics, speech and signal processing (ICASSP)","author":"Afshar","year":"2019"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1002\/14651858.CD010783.pub2","article-title":"Mini-mental state examination (MMSE) for the detection of Alzheimer's disease and other dementias in people with mild cognitive impairment (MCI)","volume":"3","author":"Arevalo-Rodriguez","year":"2015","journal-title":"Cochrane Database Syst. Rev."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"805","DOI":"10.1006\/nimg.2000.0582","article-title":"Voxel-based morphometry\u2014the methods","volume":"11","author":"Ashburner","year":"2000","journal-title":"NeuroImage"},{"key":"ref4","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/978-3-642-27772-6_298-2","article-title":"Structural and functional magnetic resonance","volume-title":"Encyclopedia of psychopharmacology","author":"Baltes","year":"2014"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"810","DOI":"10.3390\/healthcare9070810","article-title":"Smart health system to detect dementia disorders using virtual reality","volume":"9","author":"Bayahya","year":"2021","journal-title":"Healthcare"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"1953","DOI":"10.3390\/math10121953","article-title":"Older adults get lost in virtual reality: visuospatial disorder detection in dementia using a voting approach based on machine learning algorithms","volume":"10","author":"Bayahya","year":"2022","journal-title":"Mathematics"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"516","DOI":"10.3390\/rs13030516","article-title":"Vision transformers for remote sensing image classification","volume":"13","author":"Bazi","year":"2021","journal-title":"Remote Sens"},{"key":"ref8","doi-asserted-by":"crossref","DOI":"10.1109\/ICCVW.2019.00246","article-title":"Gcnet: non-local networks meet squeeze-excitation networks and beyond","volume-title":"Proceedings of the IEEE\/CVF international conference on computer vision workshops","author":"Cao","year":"2019"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"1694","DOI":"10.3390\/s23031694","article-title":"Convolution neural networks and self-attention learners for Alzheimer dementia diagnosis from brain MRI","volume":"23","author":"Carcagn\u00ec","year":"2023","journal-title":"Sensors"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"4712","DOI":"10.3390\/rs13224712","article-title":"Review of image classification algorithms based on convolutional neural networks","volume":"13","author":"Chen","year":"2021","journal-title":"Remote Sens"},{"key":"ref11","article-title":"Autoaugment: learning augmentation policies from data","author":"Cubuk","year":"2018"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1038\/s41591-018-0316-z","article-title":"A guide to deep learning in healthcare","volume":"25","author":"Esteva","year":"2019","journal-title":"Nat. Med."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1007\/s13735-017-0141-z","article-title":"A review of semantic segmentation using deep neural networks","volume":"7","author":"Guo","year":"2018","journal-title":"Int. J. Multimed. Inf. Retr."},{"key":"ref14","first-page":"770","article-title":"Deep Residual Learning for Image Recognition\u201d\u2014was presented at the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016, held from June 27 to June 30, 2016","volume-title":"Las Vegas","author":"He","year":"2016"},{"key":"ref15","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2019.00065","article-title":"Bag of tricks for image classification with convolutional neural networks","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition","author":"He","year":"2019"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"118933","DOI":"10.1016\/j.neuroimage.2022.118933","article-title":"FastSurferVINN: building resolution-independence into deep learning segmentation methods\u2014a solution for HighRes brain MRI","volume":"251","author":"Henschel","year":"2022","journal-title":"NeuroImage"},{"key":"ref17","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-80568-5_5","article-title":"Deep learning of brain asymmetry images and transfer learning for early diagnosis of dementia","volume-title":"International conference on engineering applications of neural networks","author":"Herzog","year":"2021"},{"key":"ref18","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2017.243","article-title":"Densely connected convolutional networks","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition","author":"Huang","year":"2017"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"18007","DOI":"10.1038\/s41598-022-22917-3","article-title":"Deep learning-based diagnosis of Alzheimer\u2019s disease using brain magnetic resonance images: an empirical study","volume":"12","author":"Kim","year":"2022","journal-title":"Sci. Rep."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"196","DOI":"10.1109\/TMI.2009.2035616","article-title":"elastix: A toolbox for intensity-based medical image registration","volume":"29","author":"Klein","year":"2009","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref21","article-title":"Data efficient deep learning for medical image analysis: a survey","author":"Kumari","year":"2023"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"2400044","DOI":"10.1002\/aisy.202400044","article-title":"Deformable capsules for object detection","volume":"6","author":"Lalonde","year":"2024","journal-title":"Adv. Intell. Syst."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"Lecun","year":"2015","journal-title":"Nature"},{"key":"ref24","first-page":"510","article-title":"Selective Kernel Networks\u2014was presented at the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2019, held from June 16 to June 20, 2019","volume-title":"Long Beach","author":"Li","year":"2019"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"1059","DOI":"10.1016\/j.jalz.2019.02.007","article-title":"A deep learning model for early prediction of Alzheimer's disease dementia based on hippocampal magnetic resonance imaging data","volume":"15","author":"Li","year":"2019","journal-title":"Alzheimers Dement."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"2377","DOI":"10.3390\/diagnostics13142377","article-title":"Comparative analysis of image processing techniques for enhanced MRI image quality: 3D reconstruction and segmentation using 3D U-net architecture","volume":"13","author":"Lim","year":"2023","journal-title":"Diagnostics"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med. Image Anal."},{"key":"ref28","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52688.2022.01167","article-title":"A convnet for the 2020s","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition","author":"Liu","year":"2022"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1002\/14651858.CD009628.pub2","article-title":"Structural magnetic resonance imaging for the early diagnosis of dementia due to Alzheimer's disease in people with mild cognitive impairment","volume":"3","author":"Lombardi","year":"2020","journal-title":"Cochrane Database Syst. Rev."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"1892","DOI":"10.4236\/jamp.2017.59159","article-title":"Automatic Alzheimer\u2019s disease recognition from MRI data using deep learning method","volume":"5","author":"Luo","year":"2017","journal-title":"J. Appl. Math. Phys."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"7232","DOI":"10.1002\/alz.14234","article-title":"The adni4 digital study: a novel approach to recruitment, screening, and assessment of participants for ad clinical research","volume":"20","author":"Miller","year":"2024","journal-title":"Alzheimers Dement."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"2860","DOI":"10.3390\/electronics10222860","article-title":"Multi-method analysis of medical records and MRI images for early diagnosis of dementia and Alzheimer\u2019s disease based on deep learning and hybrid methods","volume":"10","author":"Mohammed","year":"2021","journal-title":"Electronics"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"869","DOI":"10.1016\/j.nic.2005.09.008","article-title":"The Alzheimer\u2019s disease neuroimaging initiative","volume":"15","author":"Mueller","year":"2005","journal-title":"Neuroimaging Clin. N. Am."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"90319","DOI":"10.1109\/ACCESS.2021.3090474","article-title":"Demnet: a deep learning model for early diagnosis of Alzheimer diseases and dementia from MR images","volume":"9","author":"Murugan","year":"2021","journal-title":"IEEE Access"},{"key":"ref35","first-page":"580","article-title":"Detection of Alzheimer\u2019s disease (AD) in MRI images using deep learning","volume":"10","author":"Pradhan","year":"2021","journal-title":"Int. J. Eng. Res. Technol."},{"key":"ref36","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-73661-2_5","article-title":"Mobilenetv4: universal models for the mobile ecosystem","volume-title":"European conference on computer vision","author":"Qin","year":"2025"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"3404","DOI":"10.1038\/s41467-022-31037-5","article-title":"Multimodal deep learning for Alzheimer\u2019s disease dementia assessment","volume":"13","author":"Qiu","year":"2022","journal-title":"Nat. Commun."},{"key":"ref38","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-24574-4_28","article-title":"U-net: Convolutional networks for biomedical image segmentation","volume-title":"Medical image computing and computer-assisted intervention\u2013MICCAI 2015: 18th international conference, Munich, Germany, October 5\u20139, 2015, proceedings, part iii 18","author":"Ronneberger","year":"2015"},{"key":"ref39","doi-asserted-by":"crossref","DOI":"10.1109\/INOCON57975.2023.10101314","article-title":"Alzheimer\u2019s disease classification on sMRI images using convolutional neural networks and transfer learning based methods","volume-title":"2023 2nd international conference for innovation in technology (INOCON)","author":"Sai","year":"2023"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"217","DOI":"10.3390\/brainsci9090217","article-title":"A deep learning approach for diagnosis of mild cognitive impairment based on MRI images","volume":"9","author":"Gorji","year":"2019","journal-title":"Brain Sci."},{"key":"ref41","article-title":"Efficientnet: rethinking model scaling for convolutional neural networks","volume-title":"International conference on machine learning","author":"Tan","year":"2019"},{"key":"ref42","doi-asserted-by":"crossref","DOI":"10.1109\/CBMS58004.2023.00261","article-title":"Claudia: cloud-based automatic diagnosis of Alzheimer's prodromal stage and disease from 3d brain magnetic resonance","volume-title":"2023 IEEE 36th international symposium on computer-based medical systems (CBMS)","author":"Tomassini","year":"2023"},{"key":"ref43","doi-asserted-by":"publisher","first-page":"2350006","DOI":"10.4015\/S1016237223500060","article-title":"Effectiveness of learning rate in dementia severity prediction using VGG16","volume":"35","author":"Torghabeh","year":"2023","journal-title":"Biomed. Eng. Appl. Basis Commun."},{"key":"ref44","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR42600.2020.01155","article-title":"ECA-Net: efficient channel attention for deep convolutional neural networks","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition","author":"Wang","year":"2020"},{"key":"ref45","article-title":"ResNet strikes back: an improved training procedure in Timm","author":"Wightman","year":"2021"},{"key":"ref46","article-title":"Making convolutional networks shift-invariant again","volume-title":"International conference on machine learning","author":"Zhang","year":"2019"},{"key":"ref47","first-page":"3921","article-title":"Understanding deep gradient leakage via inversion influence functions","volume":"36","author":"Zhang","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref48","doi-asserted-by":"publisher","first-page":"E66367","DOI":"10.1371\/journal.pone.0066367","article-title":"MRI markers for mild cognitive impairment: comparisons between white matter integrity and gray matter volume measurements","volume":"8","author":"Zhang","year":"2013","journal-title":"PLoS One"},{"key":"ref49","doi-asserted-by":"publisher","first-page":"494","DOI":"10.3390\/rs11050494","article-title":"Remote sensing image scene classification using cnn-capsnet","volume":"11","author":"Zhang","year":"2019","journal-title":"Remote Sens"},{"key":"ref50","doi-asserted-by":"publisher","first-page":"576","DOI":"10.1002\/mp.13300","article-title":"AnatomyNet: deep learning for fast and fully automated whole-volume segmentation of head and neck anatomy","volume":"46","author":"Zhu","year":"2019","journal-title":"Med. Phys."}],"container-title":["Frontiers in Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnins.2025.1638022\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T14:46:10Z","timestamp":1764081970000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnins.2025.1638022\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,25]]},"references-count":50,"alternative-id":["10.3389\/fnins.2025.1638022"],"URL":"https:\/\/doi.org\/10.3389\/fnins.2025.1638022","relation":{},"ISSN":["1662-453X"],"issn-type":[{"value":"1662-453X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,25]]},"article-number":"1638022"}}