AGED-ViT: A Novel Transformer-Based Framework for Diagnosis of Alzheimer’s Disease by Leveraging Gene Expression Data
Bibliographic record
Abstract
Alzheimer’s disease (AD) is a growing global health concern and correct diagnosis is crucial for effective treatment. In this study, we present a novel method for the detection of AD using gene expression data from blood samples. We normalized and combined four publicly available Alzheimer’s datasets and trained a Vision Transformer (ViT) model. This combined dataset had almost seven times more features than patient samples which can cause models to overfit on the training data. To overcome this issue, we employed Linear Discriminant Analysis (LDA) to reduce the dimensionality of the data and noise injection to encourage generalizability and robustness. We then compared our model to several state-of-the-art models that used Support Vector Machines (SVMs), Convolutional Neural Networks (CNNs), and Deep Neural Networks (DNNs). Our model, AGED-ViT, achieved an average accuracy of 88.4% and area under the curve (AUC) of 0.951 on the combined dataset, outperforming previous methods. Our results demonstrate the importance of preprocessing techniques for data with more features than samples to reduce overfitting, as well as the powerful predictive capabilities of ViTs, establishing a foundation for further exploration and optimization of the transformer architecture in the context of genomic diagnosis. This study can contribute to improving the accuracy of AD diagnosis, thus facilitating intervention and leading to a more promising outcome for patients.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".