Early detection of Alzheimer’s disease progression: comparative evaluation of deep learning models
Bibliographic record
Abstract
The accurate diagnosis and monitoring of Alzheimer's disease (AD) is particularly critical given the increasing number of cases worldwide. Improving forecasting precision using deep learning models on neuroimaging biomarkers can aid in more accurately predicting Alzheimer's associated disease progression. In this work, we assess two separate 3D Convolutional Neural Network (CNN) models for binary AD progression classification based on MRIs of the brain's structure. The first model uses a whole volume approach and processes entire MRI scans, thus requiring little computational power and minimal preprocessing compared to other methods. Alternatively, the second model applies voxel-level scrutiny by examining specific pre-defined brain regions that have statistically significant grey matter volume differences from cohort analyses. MRI preprocessing includes N4 bias field correction, segmentation of tissues, alignment to the Montreal Neurological Institute (MNI) space, and Gaussian smoothing for homogenization of image quality. For the region-focused model, feature extraction is driven by neuroanatomy, concentrating on areas where AD shows shrinkage changes. The full-volume CNN achieved a 94% validation accuracy, demonstrating high computational efficiency with its simpler architecture, while the region-guided model reached 95% accuracy by leveraging more complex domain-specific structural biomarkers, highlighting enhanced performance at the cost of increased model intricacy. This study highlights the potential of combining deep learning frameworks with neuroimaging biomarkers to improve early detection and monitoring of AD. While our findings highlight the value of guided feature selection and volumetric data evaluation in improving diagnostic precision, they are derived solely from the ADNI dataset and must be validated on more diverse clinical populations.
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 | 0.007 | 0.012 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".