From longitudinal measurements to image classification: Application to longitudinal MRI in Alzheimer’s disease
Bibliographic record
Abstract
Abstract We propose a novel method of constructing representations of multiple one-dimensional longitudinal measurements as two-dimensional grey-scale images. This can be used to turn classification problems from longitudinal settings into simpler image classification problems, allowing for the application of newer deep learning methods on longitudinal measurements. Our approach is applicable to situations with balanced or imbalanced longitudinal data sets, and where there are missing data at some time points. To evaluate our approach, we apply it to an important and challenging task: the prediction of dementia from brain volume trajectories derived from longitudinal MRI. We construct an ensemble of convolutional neural network models to classify two groups of subjects: those diagnosed with mild cognitive impairment at all examinations (stable MCI) versus those starting out as MCI but later converting to Alzheimer’s disease (converted AD). Models were trained on image representations derived from N = 736 subjects sourced from the ADNI database (471/265 sMCI/cAD). We obtained an accuracy of a resulting ensemble model of 76%, measured on an independent test set. Our approach is simple and easy to apply but competitive (in terms of accuracy) with results reported in other machine learning approaches with similar classification on comparable tasks. This indicates that our approach can lead to useful representations of longitudinal data.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.005 | 0.009 |
| Research integrity | 0.000 | 0.003 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".