MedBackdoorGuard: Detecting and Mitigating Backdoor Attacks in Deep Learning Models for MRI-Based Alzheimer's Disease and Mild Cognitive Impairment Diagnosis

Topic: Brain imaging, connectivity and assessment

Deep learning classifiers trained on structural magnetic resonance imaging (MRI) are increasingly proposed for early detection of Alzheimer’s disease (AD) and mild cognitive impairment (MCI), yet the pipelines that produce these models depend on shared datasets, pre-trained weights, and federated updates that an adversary can poison. A backdoored diagnostic model behaves correctly on ordinary scans but collapses to an attacker-chosen label whenever a hidden trigger is present, which turns a prognostic instrument for cognitive aging into a silent source of misdiagnosis. This article proposes MedBackdoorGuard, a defense framework that detects and mitigates backdoor attacks in MRI-based AD and MCI classifiers without prior knowledge of the trigger. The framework couples spectral latent screening of training data with latent trigger isolation in activation space, then immunizes the model through trigger-sensitive neuron pruning and fine-tuning under an anatomical attention consistency constraint that anchors the decision on medial temporal structures. A runtime inference guard adds superimposition entropy and anatomically informed reconstruction to purify suspicious scans at deployment. Across ADNI, OASIS-1, and a public four-class dementia MRI benchmark under patch, blended, warping, and frequency triggers, MedBackdoorGuard reduces the mean attack success rate from 97.4% to 3.0% while holding clean accuracy within 0.6 percentage points of the undefended model, reaches a poisoned-sample detection AUROC of 97.4%, and identifies 59 of 60 backdoored models. Compared with the strongest defense baseline, attack success rate falls by 15.1 percentage points and clean accuracy rises by 1.7 percentage points. The framework restores MCI recall on triggered scans from 3.9% to 88.3%, which is the outcome that matters for early detection in cognitive aging.

Key findings

• Compared with the strongest defense baseline, attack success rate falls by 15.1 percentage points and clean accuracy rises by 1.7 percentage points.

New to this subject? Start with our plain-English guide: Digital health →

Study details

• Effect size97.4% to 3.0%; 97.4%; 3.9% to 88.3%
• Evidence strengthLow

Read the original

• DOI: 10.20944/preprints202609.1729.v1
• Original source →

Free copy unknown: no free-copy check has an answer for this record yet. That does not mean it is paywalled.

@article{pasha2026-30f16a55,
  author  = {Pasha, S. M. M. and Sohag, SR. and Ghanem, H.},
  title   = {{MedBackdoorGuard: Detecting and Mitigating Backdoor Attacks in Deep Learning Models for MRI-Based Alzheimer's Disease and Mild Cognitive Impairment Diagnosis}},
  journal = {Preprints.org},
  year    = {2026},
  doi     = {10.20944/preprints202609.1729.v1},
  url     = {https://doi.org/10.20944/preprints202609.1729.v1}
}

Related studies in Digital health tools

• Machine learning diagnosis of mild cognitive impairment using advanced diffusion MRI and CSF biomarkers Alzheimer s & Dementia Diagnosis Assessment & Disease Monitoring · 2025-07-01 · usefulness score 41/100
• Cortical Thickness from MRI to Predict Conversion from Mild Cognitive Impairment to Dementia in Parkinson Disease: A Machine Learning-based Model Radiology · 2021-05-25 · usefulness score 46/100
• Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset NeuroImage · 2023-03-29 · usefulness score 48/100
• Automated classification of Alzheimer's disease and mild cognitive impairment using a single MRI and deep neural networks NeuroImage. Clinical · 2018-12-18 · usefulness score 48/100
• Developing a machine learning model for detecting depression, anxiety, and apathy in older adults with mild cognitive impairment using speech and facial expressions: A cross-sectional observational study International journal of nursing studies · 2023-07-07 · usefulness score 52/100
• Plasma d-glutamate levels for detecting mild cognitive impairment and Alzheimer's disease: Machine learning approaches Journal of psychopharmacology (Oxford, England) · 2021-02-15 · usefulness score 47/100

Also tagged “convolutional”

205 studies in the library carry the topic tag “convolutional”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 205 tagged studies.

• AT(N)-Anchored Multimodal Deep Learning for Alzheimer’s Staging: A Taxonomy and Continuous Burden Index for Low-Resource Settings Preprints.org · 2026-09-25 · usefulness score 16/100 · Tau biomarkers
• Intelligence Breaks the System It Controls Research Square · 2026-08-27 · usefulness score 16/100 · Long-term cohort studies
• Neuron-level explainable AI for Alzheimer's Disease assessment from fundus images Scientific reports · 2024-04-02 · usefulness score 59/100 · Brain scans and imaging

Similar studies

• DEMNET: A Deep Learning Model for Early Diagnosis of Alzheimer Diseases and Dementia From MR Images 2021 · Long-term cohort studies
• Brain MRI analysis for Alzheimer's disease diagnosis using an ensemble system of deep convolutional neural networks 2018 · Digital health tools
• Multimodal deep learning models for early detection of Alzheimer's disease stage 2021 · Digital health tools
• Convolutional neural networks for classification of Alzheimer's disease: Overview and reproducible evaluation 2020 · Digital health tools
• MRI signatures of brain age and disease over the lifespan based on a deep brain network and 14 468 individuals worldwide 2020 · Digital health tools
• Convolutional Neural Networks-Based MRI Image Analysis for the Alzheimer's Disease Prediction From Mild Cognitive Impairment 2018 · Digital health tools

More on Digital health tools

• A review that covers this ground Systematic review: fluid biomarkers and machine learning methods to improve the diagnosis from mild cognitive impairment to Alzheimer's disease Alzheimer's research & therapy · 2023-10-14 · usefulness score 63/100
• Newest in this subject Speaker-Wise Dementia Screening fromNaturalistic Speech with Lexical Entropy, Semantic Drift, and Multimodal Fusion Research Square · 2026-10-21 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Integrating Multisource Target Evidence with Machine-Learning Potency Prediction for Alzheimer’s Disease Drug Discovery bioRxiv · 2026-10-20 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Digital health tools →