Development of an Electronic Medical Record–Based Score for Heart Failure Prediction in Cancer Survivors
Bibliographic record
Abstract
BACKGROUND: Awareness of heart failure (HF) as a long-term complication of cancer has led to an interest in HF surveillance among survivors. However, existing HF risk scores are not tailored for survivors and not designed for the use in electronic medical records (EMRs) or administrative data sets where clinical data such as blood pressure and pathology results are often unavailable. OBJECTIVES: The objective of the study is to develop a cancer-specific incident HF risk score suitable for screening in EMR or administrative data sets. METHODS: The cancer-specific HF prediction from EMRs in survivor health care (CHERISH) risk score was developed from risk variables identified in 16,191 cancer survivors (mean 61 years; 59.5% female) derived from the UK Biobank. External validation was conducted in a population-based Ontario cohort (n = 446,096; mean 67 years; 53.9% female). HF risk classification with CHERISH was compared against the ARIC (Atherosclerotic Risk In Community)-HF score using area under the curve (AUC). RESULTS: The CHERISH score incorporates 11 clinical variables-age, years since cancer diagnosis, coronary heart disease, arrhythmia, myocardial infarct, diabetes, hypertension, leukemia, non-Hodgkin lymphoma, lung cancer, and breast cancer. CHERISH demonstrated strong prediction of 10-year HF incidence during internal validation (AUC: 0.829), exceeding ARIC-HF (AUC: 0.697; P < 0.001). In external validation, CHERISH showed an AUC of 0.721 in predicting 10-year HF incidence, compared to an AUC of 0.751 (P < 0.001) with ARIC-HF. CONCLUSIONS: The integration of the CHERISH score into EMR systems may provide large-scale, automated HF risk assessment in cancer survivors, using routinely collected clinical 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 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 | 0.004 | 0.013 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 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".