{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T20:20:51Z","timestamp":1790799651705,"version":"4.1.0"},"reference-count":37,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T00:00:00Z","timestamp":1772668800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJERPH"],"abstract":"<jats:p>Alzheimer\u2019s disease (AD), the most common form of dementia, is a progressive and irreversible neurodegenerative disorder. Structural MRI is widely used for diagnosis, revealing brain changes associated with AD. However, these alterations are often subtle and difficult to detect manually, particularly at early stages. Early intervention during prodromal stages, such as mild cognitive impairment (MCI), can help slow disease progression, highlighting the need for reliable automated methods. In this work, we introduce a dual-level evaluation framework comparing fifteen deep learning architectures, including convolutional neural networks (CNNs), Transformers, and hybrid models, for classifying AD, MCI, and cognitively normal (CN) subjects using the ADNI dataset. A central focus of our work is the impact of robust and standardized preprocessing pipelines, which we identified as a critical yet underexplored factor influencing model reliability. By evaluating performance at both slice-level and scan-level, we reveal that multi-slice aggregation affects architectures asymmetrically. By systematically optimizing preprocessing steps to reduce data variability and enhance feature consistency, we established preprocessing quality as an essential determinant of deep learning performance in neuroimaging. Experimental results show that CNNs and hybrid pre-trained models outperform Transformer-based models in both slice-level and scan-level classification. ConvNeXtV2-L achieved the best scan-level performance (91.07%), EfficientNetV2-L the highest slice-level accuracy (86.84%), and VGG19 balanced results (86.07%\/88.52%). ConvNeXtV2-L and SwinV1-L exhibited scan-level improvements of 7.60% and 9.04% respectively, while EfficientNetV2-L experienced degradation of 2.66%, demonstrating that architectural selection and aggregation strategy are interdependent factors. These findings suggest that carefully designed preprocessing not only improves classification accuracy but may also serve as a foundation for more reproducible and interpretable Alzheimer\u2019s disease detection pipelines.<\/jats:p>","DOI":"10.3390\/ijerph23030322","type":"journal-article","created":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T11:52:11Z","timestamp":1772711531000},"page":"322","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Multi-Architecture Deep Learning for Early Alzheimer\u2019s Detection in MRI: Slice- and Scan-Level Analysis"],"prefix":"10.3390","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-8190-3010","authenticated-orcid":false,"given":"Isabelle","family":"Bricaud","sequence":"first","affiliation":[{"name":"School of Computing, University of Kent, Canterbury CT2 7PE, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6734-9424","authenticated-orcid":false,"given":"Giovanni Luca","family":"Masala","sequence":"additional","affiliation":[{"name":"School of Computing, University of Kent, Canterbury CT2 7PE, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,5]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"3767","DOI":"10.1007\/s11042-023-15738-7","article-title":"A deep learning framework for early diagnosis of Alzheimer\u2019s disease on MRI images","volume":"83","author":"Arafa","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref_2","unstructured":"World Health Organization (2025, November 10). Dementia: Key Facts. Available online: https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/dementia."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"3463","DOI":"10.1038\/s41598-024-53733-6","article-title":"A novel CNN architecture for accurate early detection and classification of Alzheimer\u2019s disease using MRI data","volume":"14","author":"Amer","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Alzheimer\u2019s Association (2023). 2023 Alzheimer\u2019s Disease Facts and Figures. Alzheimers Dement., 19, 1598\u20131695.","DOI":"10.1002\/alz.13016"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"306","DOI":"10.1097\/WAD.0b013e3181a6bebc","article-title":"Missed and Delayed Diagnosis of Dementia in Primary Care: Prevalence and Contributing Factors","volume":"23","author":"Bradford","year":"2009","journal-title":"Alzheimer Dis. Assoc. Disord."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"280","DOI":"10.1016\/j.jalz.2011.03.003","article-title":"Toward Defining the Preclinical Stages of Alzheimer\u2019s Disease: Recommendations from the National Institute on Aging-Alzheimer\u2019s Association Workgroups on Diagnostic Guidelines for Alzheimer\u2019s Disease","volume":"7","author":"Sperling","year":"2011","journal-title":"Alzheimers Dement."},{"key":"ref_7","first-page":"178","article-title":"A Deep Learning-Based Technique to Determine Various Stages of Alzheimer\u2019s Disease from 3D Brain MRI Images","volume":"Volume 14416","year":"2023","journal-title":"Information Integration and Web Intelligence"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Lim, B.Y., Lai, K.W., Haiskin, K., Kulathilake, K.A.S.H., Ong, Z.C., Hum, Y.C., Dhanalakshmi, S., Wu, X., and Zuo, X. (2022). Deep Learning Model for Prediction of Progressive Mild Cognitive Impairment to Alzheimer\u2019s Disease Using Structural MRI. Front. Aging Neurosci., 14.","DOI":"10.3389\/fnagi.2022.876202"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e12050","DOI":"10.1002\/trc2.12050","article-title":"Alzheimer\u2019s Disease Drug Development Pipeline: 2020","volume":"6","author":"Cummings","year":"2020","journal-title":"Alzheimers Dement. Transl. Res. Clin. Interv."},{"key":"ref_10","first-page":"281","article-title":"Automated Classification of Alzheimer\u2019s Disease Stages Using T1-Weighted sMRI Images and Machine Learning","volume":"Volume 364","author":"Jha","year":"2024","journal-title":"Proceedings of Congress on Control, Robotics, and Mechatronics"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"104","DOI":"10.1186\/s13195-021-00837-0","article-title":"Predict Alzheimer\u2019s disease using hippocampus MRI data: A lightweight 3D deep convolutional network model with visual and global shape representations","volume":"13","author":"Katabathula","year":"2021","journal-title":"Alzheimers Res. Ther."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1227","DOI":"10.1212\/01.wnl.0000180958.22678.91","article-title":"Brain Atrophy Rates Predict Subsequent Clinical Conversion in Normal Elderly and Amnestic MCI","volume":"65","author":"Jack","year":"2005","journal-title":"Neurology"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Abbasian, P., and Hammond, T.A. (2024). A Comprehensive Analysis of Early Alzheimer Disease Detection from 3D sMRI Images Using Deep Learning Frameworks. Information, 15.","DOI":"10.3390\/info15120746"},{"key":"ref_14","first-page":"219","article-title":"Classification on Alzheimer\u2019s Disease MRI Images with VGG-16 and VGG-19","volume":"Volume 312","author":"Choudrie","year":"2023","journal-title":"IOT with Smart Systems"},{"key":"ref_15","unstructured":"Alzheimer\u2019s Disease Neuroimaging Initiative (ADNI) (2025, May 22). ADNI Database. Available online: http:\/\/adni.loni.usc.edu\/."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Lowekamp, B.C., Chen, D.T., Ib\u00e1\u00f1ez, L., and Blezek, D. (2013). The Design of SimpleITK. Front. Neuroinform., 7.","DOI":"10.3389\/fninf.2013.00045"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"193","DOI":"10.1007\/s12021-024-09655-9","article-title":"DeepN4: Learning N4ITK Bias Field Correction for T1-weighted Images","volume":"22","author":"Kanakaraj","year":"2024","journal-title":"Neuroinformatics"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1310","DOI":"10.1109\/TMI.2010.2046908","article-title":"N4ITK: Improved N3 bias correction","volume":"29","author":"Tustison","year":"2010","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_19","first-page":"s13742-016-0147-0-w","article-title":"DIPY: Brain tissue classification","volume":"5","author":"Garyfallidis","year":"2016","journal-title":"GigaScience"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Garyfallidis, E., Brett, M., Amirbekian, B., Rokem, A., Van Der Walt, S., Descoteaux, M., and Nimmo-Smith, I. (2014). Dipy, a library for the analysis of diffusion MRI data. Front. Neuroinform., 8.","DOI":"10.3389\/fninf.2014.00008"},{"key":"ref_21","first-page":"1293","article-title":"A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM)","volume":"356","author":"Mazziotta","year":"2001","journal-title":"Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"S102","DOI":"10.1016\/S1053-8119(09)70884-5","article-title":"Unbiased nonlinear average age-appropriate brain templates from birth to adulthood","volume":"47","author":"Fonov","year":"2009","journal-title":"NeuroImage"},{"key":"ref_23","unstructured":"Mattes, D., Haynor, D.R., Vesselle, H., Lewellen, T.K., and Eubank, W. (2001, January 17\u201322). Nonrigid multimodality image registration. Proceedings of the Medical Imaging 2001: Image Processing, San Diego, CA, USA."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"59383","DOI":"10.1109\/ACCESS.2023.3285115","article-title":"Diagnosis of Alzheimer\u2019s Disease Using Convolutional Neural Network With Select Slices by Landmark on Hippocampus in MRI Images","volume":"11","author":"Pusparani","year":"2023","journal-title":"IEEE Access"},{"key":"ref_25","unstructured":"Simonyan, K., and Zisserman, A. (2015). Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Nafis, F.A., Akib, N., Hossain, M., Farasha, M.Z., Jobayer, M., and Shawon, M.M.H. (2024, January 28\u201329). CNN and Transfer Learning-based Deep Learning Architectures for Alzheimer\u2019s Disease Detection from MRI Scan: A Comparative Analysis. Proceedings of the 2024 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON), Dhaka, Bangladesh.","DOI":"10.1109\/BECITHCON64160.2024.10962572"},{"key":"ref_27","unstructured":"Tan, M., and Le, Q.V. (2021, January 18\u201324). EfficientNetV2: Smaller Models and Faster Training. Proceedings of the 38th International Conference on Machine Learning (ICML), Virtual."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"181","DOI":"10.37934\/araset.34.3.181191","article-title":"Transfer Learning for Alzheimer\u2019s Disease Diagnosis Using EfficientNet-B0 Convolutional Neural Network","volume":"35","author":"Wong","year":"2023","journal-title":"J. Adv. Res. Appl. Sci. Eng. Technol."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Woo, S., Debnath, S., Hu, R., Chen, X., Liu, Z., Kweon, I.S., and Xie, S. (2023, January 17\u201324). ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders. Proceedings of the 2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.01548"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Miller, C., Portlock, T., Nyaga, D.M., and O\u2019Sullivan, J.M. (2024). A review of model evaluation metrics for machine learning in genetics and genomics. Front. Bioinform., 4.","DOI":"10.3389\/fbinf.2024.1457619"},{"key":"ref_31","first-page":"100782","article-title":"Automated Alzheimer\u2019s disease classification using deep learning models with Soft-NMS and improved ResNet50 integration","volume":"17","author":"Chen","year":"2024","journal-title":"J. Radiat. Res. Appl. Sci."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"974","DOI":"10.1016\/j.trci.2019.10.001","article-title":"Convolution neural network\u2013based Alzheimer\u2019s disease classification using hybrid enhanced independent component analysis based segmented gray matter of T2 weighted magnetic resonance imaging with clinical valuation","volume":"5","author":"Basheera","year":"2019","journal-title":"Alzheimers Dement. Transl. Res. Clin. Interv."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Sahumbaiev, I., Popov, A., Ram\u00edrez, J., G\u00f3rriz, J.M., and Ortiz, A. (2018, January 10\u201317). 3D-CNN HadNet classification of MRI for Alzheimer\u2019s Disease diagnosis. Proceedings of the 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS\/MIC), Sydney, NSW, Australia.","DOI":"10.1109\/NSSMIC.2018.8824317"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"2201","DOI":"10.1007\/s13369-021-06131-3","article-title":"Detecting the Stages of Alzheimer\u2019s Disease with Pre-trained Deep Learning Architectures","volume":"47","year":"2022","journal-title":"Arab. J. Sci. Eng."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"14487","DOI":"10.1007\/s00521-022-07263-9","article-title":"Two-stage deep learning model for Alzheimer\u2019s disease detection and prediction of the mild cognitive impairment time","volume":"34","author":"Saleh","year":"2022","journal-title":"Neural Comput. Appl."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"105032","DOI":"10.1016\/j.compbiomed.2021.105032","article-title":"Deep learning based pipelines for Alzheimer\u2019s disease diagnosis: A comparative study and a novel deep-ensemble method","volume":"141","author":"Loddo","year":"2022","journal-title":"Comput. Biol. Med."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Fei-Fei, L. (2009, January 20\u201325). ImageNet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.","DOI":"10.1109\/CVPR.2009.5206848"}],"container-title":["International Journal of Environmental Research and Public Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1660-4601\/23\/3\/322\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T11:55:23Z","timestamp":1772711723000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1660-4601\/23\/3\/322"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,5]]},"references-count":37,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["ijerph23030322"],"URL":"https:\/\/doi.org\/10.3390\/ijerph23030322","relation":{},"ISSN":["1660-4601"],"issn-type":[{"value":"1660-4601","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,5]]}}}