Differential Treatment Benefit Prediction For Treatment Selection in Depression: A Deep Learning Analysis of STAR*D and CO-MED Data
Bibliographic record
Abstract
Abstract Background Depression affects one in nine people, but treatment response rates remain low. There is significant potential in the use of computational modelling techniques to predict individual patient responses and thus provide more personalized treatment. Deep learning is a promising computational technique that can be used for differential treatment selection based on predicted remission probability. Methods Using STAR*D and CO-MED trial data, we employed deep neural networks to predict remission after feature selection. Differential treatment benefit was estimated in terms of improvement of population remission rates after application of the model for treatment selection using both naive and conservative approaches. The naïve approach assessed population remission rate in five sets of 200 patients held apart from the training set; the conservative approach used bootstrapping for sample generation and focused on population remission rate for patients who actually received the drug predicted by the model compared to the general population. Results Our deep learning model predicted remission in a pooled CO-MED/STAR*D dataset (including four treatments) with an AUC of 0.69 using 17 input features. Our naive analysis showed an improvement of remission of over 30% (from a 34.33% population remission rate to 46.12%). Our conservative analysis showed a 7.2% improvement in population remission rate (p= 0.01, C.I. 2.48% ± .5%). Conclusion Our model serves as proof-of-concept that deep learning has utility in differential prediction of antidepressant response when selecting from a number of treatment options. These models may have significant real-world clinical implications.
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 | 0.006 | 0.011 |
| Meta-epidemiology (narrow) | 0.000 | 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 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".