Alzheimer’s related dementia severity classification from magnetic resonance imaging using derivative-free optimization of convolutional neural network

Topic: Amyloid, tau and blood-test diagnosis

Magnetic resonance imaging based dementia severity assessment facilitates timely diagnosis of Alzheimer’s disease (AD) progression. While convolutional neural networks (CNNs) have shown effectiveness in classifying AD stages, their performance can be limited by high computational demands and class imbalance in training data. The article introduces a derivative-free optimization (DFO) approach by integrating evolutionary algorithms, Bayesian optimization, simulated annealing, neural architecture search, and pruning techniques to optimize both trainable parameters and network topology to design a lightweight CNN model (DAPA-CNN) for classifying brain MRI images into four Alzheimer-related Dementia stages (ARDS): non-Dementia (ND), mild Dementia (MD), very mild Dementia (VMD) and moderate Dementia (MoD). The proposed framework balances the dataset using Tomek links and deep SMOTE, and enhances model interpretability with class activation maps (CAMs). DAPA-CNN achieved an accuracy of 99.59% on the Alzheimer’s disease dataset (ADD), along with balanced precision (99.60%), sensitivity (99.66%), specificity (99.87%), and F1score (99.63%). across all classes. All class-wise dice and Jaccard indices and correlation metrics (Matthews correlation coefficient and Cohen’s Kappa) exceeded 0.99. Compared to a baseline CNN and contemporary architectures, DAPA-CNN reduces the number of parameters by 85.6%, processing time by 42.8%, and memory usage by 76.6%, making it suitable for resource-constrained clinical environments.

@article{ganesan2026-41723164,
  author  = {Ganesan, Saravana Kumar and Velusamy, Parthasarathy and Parthsarathy, Praveen and Vasudevan, Asokan},
  title   = {{Alzheimer’s related dementia severity classification from magnetic resonance imaging using derivative-free optimization of convolutional neural network}},
  journal = {Scientific Reports},
  year    = {2026},
  doi     = {10.1038/s41598-026-36037-9},
  pmid    = {41723164},
  url     = {https://doi.org/10.1038/s41598-026-36037-9}
}

Key findings

• Compared to a baseline CNN and contemporary architectures, DAPA-CNN reduces the number of parameters by 85.6%, processing time by 42.8%, and memory usage by 76.6%, making it suitable for resource-constrained clinical environments.

Why it matters

This study describes a computer model that classifies MRI scans into four dementia stages with high reported accuracy while using fewer parameters and less memory than other models. It is a technical classification study and does not test any treatment, cure, or prevention for Alzheimer's disease.

New to this subject? Start with our plain-English guide: Getting a diagnosis →

Study details

• Effect size99.59%; 99.60%; 99.66%
• Evidence strengthLow

Read the original

• DOI: 10.1038/s41598-026-36037-9
• PubMed: 41723164
• Original source →
• Free full text: Scientific Reports →

Full text: free to read — OpenAlex lists this copy as free to read: the link above (checked 6 Feb 1020).

Related studies in Getting a diagnosis

• Enhancing Mild Cognitive Impairment Auxiliary Identification Through Multimodal Cognitive Assessment with Eye Tracking and Convolutional Neural Network Analysis Biomedicines · 2025-03-18 · usefulness score 41/100
• Use of structural magnetic resonance imaging to predict who will get Alzheimer's disease Annals of neurology · 2000-04-01 · usefulness score 59/100
• White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer's disease, vascular dementia, and normal aging Journal of neurology, neurosurgery, and psychiatry · 1999-07-01 · usefulness score 52/100
• Imaging of onset and progression of Alzheimer's disease with voxel-compression mapping of serial magnetic resonance images Lancet (London, England) · 2001-07-01 · usefulness score 48/100
• Assessment of cerebral blood flow in Alzheimer's disease by spin-labeled magnetic resonance imaging Annals of neurology · 2000-01-01 · usefulness score 47/100
• Incidental subcortical lesions identified on magnetic resonance imaging in the elderly. I. Correlation with age and cerebrovascular risk factors Stroke · 1986-11-01 · usefulness score 45/100

Also tagged “classification”

835 studies in the library carry the topic tag “classification”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 835 tagged studies.

• Local amygdala structural differences with 3T MRI in patients with Alzheimer disease Neurology · 2011-02-01 · usefulness score 50/100 · Brain scans and imaging
• Staging β-Amyloid Pathology With Amyloid Positron Emission Tomography JAMA neurology · 2019-11-01 · usefulness score 67/100 · Biomarkers
• Antibody-Based In Vivo PET Imaging Detects Amyloid-β Reduction in Alzheimer Transgenic Mice After BACE-1 Inhibition Journal of nuclear medicine : official publication, Society of Nuclear Medicine · 2018-05-31 · usefulness score 42/100 · Biomarkers

Similar studies

• Classifying and diagnosing Alzheimer’s disease with deep learning using 6735 brain MRI images 2025 · Brain scans and imaging
• Advanced convolutional neural network with attention mechanism for Alzheimer's disease classification using MRI 2025 · Brain scans and imaging
• Prediction and classification of Alzheimer disease based on quantification of MRI deformation 2017 · Brain scans and imaging
• Detection and Classification of Alzheimer’s Disease Using Deep and Machine Learning 2025 · Digital health tools
• Convolutional Neural Network Achieves 97.67 Percent Accuracy for Alzheimer MRI Classification 2026 · Brain scans and imaging
• Predicting Prodromal Alzheimer's Disease in Subjects with Mild Cognitive Impairment Using Machine Learning Classification of Multimodal Multicenter Diffusion-Tensor and Magnetic Resonance Imaging Data 2015 · Digital health tools

More on Getting a diagnosis

• A review that covers this ground Functional network disruption in the degenerative dementias The Lancet. Neurology · 2011-07-21 · usefulness score 54/100
• Newest in this subject The Contribution of Routinely Collected Clinical Data to Dementia Identification in Primary Care: A Cross-Sectional Study Research Square · 2026-10-28 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Beyond “Denial”: Understanding and Responding to Help-Refusal in Dementia Preprints.org · 2026-10-26 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Getting a diagnosis →