Predicting In-Hospital Cardiac Arrest Using Machine Learning Models: Protocol for a Scoping Review
Bibliographic record
Abstract
BACKGROUND: In-hospital cardiac arrest (IHCA) remains a public health conundrum with high morbidity and mortality rates. While early identification of high-risk patients could enable preventive interventions and improve survival, evidence on the effectiveness of current prediction methods remains inconclusive. Limited research exists on patients' prearrest pathophysiological status and predictive and prognostic factors of IHCA, highlighting the need for a comprehensive synthesis of predictive methodologies. OBJECTIVE: This scoping review aims to synthesize and critically evaluate the quality and quantity of clinical features and machine learning (ML) models for predicting IHCA. The review will evaluate temporal characteristics, predictive and prognostic values of prearrest clinical features, and model performance metrics. METHODS: This scoping review follows the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines and aims to synthesize studies that used ML algorithms to predict IHCA published between April 2009 and April 2024. We will conduct a comprehensive search using 4 major databases: PubMed, Web of Science, IEEE Xplore, and Embase. The inclusion criteria are peer-reviewed, English-language studies that explore ML applications for predicting IHCA in adult patients (aged ≥18 years). Exclusion criteria include review articles, preprints, non-English-language studies, and studies without specific ML metrics for IHCA prediction. Two independent reviewers will conduct the screening and data extraction using Rayyan for deduplication and ensuring study eligibility. Descriptive statistics will be used to summarize the data, and a narrative synthesis will provide insights into the clinical features used in the models, the performance metrics, and any gaps in the literature. RESULTS: A total of 2479 records were identified between April 2009-April 2024. After removing duplicates and conducting screening, 16 studies have been included in the review. Data extraction and synthesis are ongoing and are expected to be completed by June 2025. The anticipated results from this review will provide a comprehensive overview of the clinical predictors of IHCA used in ML models, including commonly reported clinical features such as vital signs, biomarkers, and comorbidities. We expect to highlight variations in data quality and quantity across studies, which may influence model performance. CONCLUSIONS: This study will contribute to advancing ML applications for IHCA prediction by addressing data challenges and promoting standardization to improve the clinical decision-making process. The results of this review are expected to inform future studies; promote consistency in the reporting of clinical features; and, ultimately, enhance the decision-making process in clinical settings, potentially leading to better outcomes for patients experiencing IHCA. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/69716.
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 | 0.073 | 0.090 |
| Meta-epidemiology (narrow) | 0.005 | 0.006 |
| Meta-epidemiology (broad) | 0.014 | 0.016 |
| Bibliometrics | 0.015 | 0.014 |
| Science and technology studies | 0.004 | 0.004 |
| Scholarly communication | 0.007 | 0.009 |
| Open science | 0.005 | 0.007 |
| Research integrity | 0.007 | 0.007 |
| Insufficient payload (model declined to judge) | 0.101 | 0.017 |
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".