Using large language models as a scalable mental status evaluation technique
Bibliographic record
Abstract
Mental health care faces a significant gap in service availability, with demand for services significantly surpassing available care. As such, building scalable and objective measurement tools for mental health evaluation is of primary concern. Given the usage of spoken language in diagnostics and treatment, it stands out as a potential methodology. With a substantial mismatch between the demand for services and the availability of care, this study focuses on leveraging large language models to bridge this gap. Here, a RoBERTa-based transformer model is fine-tuned for mental health status evaluation using natural language processing. The model analyzes written language without access to prosodic, motor, or visual cues commonly used in clinical mental status exams. Using non-clinical data from online forums and clinical data from a board-reviewed online psychotherapy trial, this study provides preliminary evidence that large language models can support symptom identification in classifying sentences with an accuracy comparable to human experts. The text dataset is expanded through augmentation using backtranslation and the model performance is optimized through hyperparameter tuning. Specifically, a RoBERTa-based model is fine-tuned on psychotherapy session text to predict whether individual sentences are symptomatic of anxiety or depression with prediction accuracy on par with clinical evaluations at 74%.
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 | 0.003 | 0.011 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".