0714 Machine learning model for Predicting phenoconversion in patients with Rem behavior disorder using clinical markers
Bibliographic record
Abstract
Abstract Introduction Idiopathic Rapid eye movement (REM) sleep behavior disorder is a condition that can be an early sign of alpha-synuclein-mediated neurodegenerative diseases, and the course of the disease can vary greatly from patient to patient. It is important to identify patients who are at risk of developing neurodegenerative diseases in the future for the purpose of future clinical trials and for patients to plan their lives accordingly. Previous research has identified various risk factors for phenoconversion in RBD patients, but these studies are not practical for use in clinical settings due to resource availability or the rarity of certain features. Additionally, most of these studies have been conducted on non-Asian populations, which may have different genetic backgrounds than Asian populations. This study aimed to develop a machine learning model to predict survival in RBD patients using clinical features commonly available in routine clinical settings. Methods This study recruited patients diagnosed with RBD based on polysomnography results and collected 34 features for each patient. Missing data were imputed and various models were applied to the data to improve performance. The model's predictive performance was evaluated using an integrated Brier score and the concordance index. Mean performance indicators were calculated from 5-fold cross-validation results. A web application hosting the final prediction model was developed and deployed on a server for use by physicians or patients. Results 173 patients were included in the study. We used the likelihood ratio test to calculate the p-values of all variables and selected the following 8 variables with p-values less than 0.1: UPDRS part III, age, history of antidepressant use, history of alcohol use, MoCA (Montreal Cognitive Assessment), PSQI-TST (Pittsburgh Sleep Quality Index - total sleep time), AHI-REM (apnea-hypopnea index - REM sleep), and education level. The random survival forest model had the best mean IBS of 0.07 and the best C-index of 0.93 Conclusion We showed that it is possible for a machine learning model to predict phenoconversion in patients with RBD using features that are commonly available in routine clinical settings Support (if any)
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 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".