Clinical implementation of an AI-enabled ECG for hypertrophic cardiomyopathy detection
Bibliographic record
Abstract
BACKGROUND: Hypertrophic cardiomyopathy (HCM) is often underdiagnosed. Artificial intelligence (AI)-based notification of HCM suspicion on a 12-lead ECG has been proposed to assist patient identification and evaluation. However, there has been no study to date to assess clinical implementation of this approach. METHODS: In an open-label, multicentre prospective cohort study, Viz HCM (Viz.ai)-an AI-ECG software alerting of suspected HCM-was implemented at five healthcare systems between January and December 2023 to identify patients >18 years of age without prior HCM diagnosis. The coprimary endpoints were the percentage of HCM-suspected cases viewed by users and the types of follow-up actions. Additional outcome measures included the time to follow-up, demographic characteristics of enrolled patients and follow-up outcomes. RESULTS: Out of 145 848 patients screened with algorithm-compliant ECGs, 4348 (3%) were alerted for suspected HCM. Users viewed 69% (3017/4348) of AI-suspected HCM cases. 217 patients met the study criteria and were enrolled with broad representation across racial and ethnic groups-including 23% Black, 9% Asian and 12% Hispanic or Latino. Of the enrolled patients, 182 (84%) had an indication for a total of 243 follow-up actions. The median (interquartile) time from ECG to diagnostic imaging indicating HCM was 7.5 (1.0-37.2) days. From the 217 enrolled patients, 17 (7.8%) were newly diagnosed with HCM-8 inpatient and 9 outpatient. During the study, deployment of an optimised algorithm operating point helped reduce the alert percentage of algorithm-screened patients from 4.4% (2097/47868) to 2.3% (2251/97980), p<0.0001, with no difference in the enrolment rate by alerts reviewed. CONCLUSION: An AI-based ECG device for HCM can be implemented successfully in a variety of clinical workflows to help identify new patients with HCM. Future study is warranted to assess scalability and comparisons to standard of care.
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".