Abstract 13225: Frailty Screening at Scale Using Core Clinical Data and Supervised Machine Learning
Bibliographic record
Abstract
Introduction: Frailty is a major risk factor for adverse health events in older adults with cardiovascular disease. Hypothesis: We sought to develop and validate a predictive model leveraging data available in the electronic health record to screen for frailty as defined by a prospective reference standard. Methods: We conducted a population-based cohort study using data from the Canadian Longitudinal Study of Aging (CLSA). From 2010-2015, the CLSA enlisted a diverse and multi-ethnic sample of community-dwelling adults 45-85 years of age. Comprehensive phenotyping was performed through interviews at participants’ homes and assessments at data collection sites. Frailty was quantified by the 47-item Frailty Index (FI) Examination, consisting of tests for age-related deficits in physical performance, body composition, cardiovascular and pulmonary physiology, cognitive and sensory function. After dividing our sample into training (80%) and test (20%) sets, we compared machine learning and linear regression models to predict the FI based on age, sex, comorbidities, and blood test results. We used the H2O AutoML platform (DAI 1.10.2) to iteratively determine the optimal model. Results: The cohort consisted of 30,097 adults with a mean age of 63±10 years and 51% females. The mean FI score was 0.28±0.08 (best-worst 0.07-0.70). The gradient-boosted machine learning model achieved an R 2 of 0.512 and a root mean squared error (RMSE) of 0.058 to regress FI scores, and an area under the curve of 0.766 to classify FI quintiles, selecting the following 10 variables in descending order of importance: age, chronic heart failure, hypertension, glycated hemoglobin, high-sensitivity C-reactive protein, high-density lipoprotein, red cell distribution width, vitamin D, female sex, and diabetes. The linear regression model achieved a similar R 2 and RMSE with all 62 input variables and a modest decline after selecting the top 10 input variables using a leaps-and-bounds algorithm. Conclusions: Frailty screening can be performed at scale for clinical or research purposes using common clinical and biochemical inputs. Our machine learning model predicted frailty with a manageable number of inputs and yielded pathophysiological insights about their relative importance.
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 | 0.009 | 0.020 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".