An adaptive learning framework for Alzheimer’s disease diagnosis using structural Magnetic Resonance Imaging data analytics
Bibliographic record
Abstract
Early and accurate diagnosis of Alzheimer’s disease (AD) is crucial for managing the disease and selecting therapies to slow its progression. Machine learning (ML) has demonstrated significant potential in improving its diagnostic accuracy with structural Magnetic Resonance Imaging (sMRI) data. However, developing robust ML models faces the challenge of domain shift in sMRI data caused by differences between datasets sources. These differences can lead to performance degradation when applying the ML models trained on one dataset to another. To address this issue, we propose a cross-domain learning framework tailored to classify AD, mild cognitive impairment (MCI), and cognitively normal (CN) subjects from sMRI data accounting for domain shift. Our approach begins by transfer-learning with pre-trained 3D ResNet50 using labeled images from the source domain. We then enhance the model through adversarial training using both source images and unlabeled target images and use the maximum mean discrepancy (MMD) to align the feature distributions across different domains at the same time. Building upon the adversarially trained model, we introduce a self-supervised learning stage with a teacher–student framework incorporated, which reduces class imbalance via dynamic class weights and reinforces domain alignment via MMD. Our proposed framework is validated to outperform existing domain adaptation approaches on 3T and 1.5T sMRI scans from Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. Notably for MCI vs.CN classification, our model achieves a high enough accuracy of 86.86% to enable early detection of dementia.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".