A Wearable Infrared Sensor for Detecting Non-ST Segment Elevation Acute Coronary Syndromes
Sengupta, P. P.; Jamthikar, A. D.; Yanamala, N.; Maganti, K.; Titus, J.; Bhavnani, S.; Sengupta, S.
Show abstract
Non-ST-segment elevation acute coronary syndrome (NSTE-ACS) is conventionally diagnosed using electrocardiography and serial blood biomarker measurements. We investigated a non-invasive, bloodless, and electrode-free diagnostic strategy using a wrist-worn infrared spectrophotometric biosensor (Infrasensor). In a prospective, multicenter study of 595 patients with suspected NSTE-ACS enrolled across 13 sites in two countries, participants were stratified into five analytical cohorts. With 200 multi-ethnic controls and a leave-one-cohort-out external validation, a machine learning model detected high-grade coronary obstruction with an area under the receiver operating characteristic curve (AUC) of 0.87 (95% CI: 0.84-0.90), 90% specificity, and 84% positive predictive value--surpassing standard risk scores. A secondary model predicted freedom from NSTE-ACS and adverse outcomes over 30 days with an AUC of 0.89 (95% CI: 0.87-0.92), 99% sensitivity, and 96% negative predictive value. These findings demonstrate the potential of the Infrasensor as a rapid, scalable point-of-care tool for early risk stratification in NSTE-ACS.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- High-sensitivity cardiac troponin on presentation to rule out myocardial infarction: a stepped-wedge cluster randomised controlled trial 94%
- Agrin promotes coordinated therapeutic processes leading to improved cardiac repair in pigs 94%
- Arrhythmia and Survival Outcomes among Black and White Patients with a Primary Prevention Defibrillator 94%
Similar papers in this journal
- Development and validation of imaging-free myocardial fibrosis prediction models, association with outcomes, and sample size estimation for phase 3 trials 95%
- The HeartMagic prospective observational study protocol - characterizing subtypes of heart failure with preserved ejection fraction 95%
- Artificial intelligence of arterial Doppler waveforms to predict major adverse outcomes among patients evaluated for peripheral artery disease 95%
Similar papers in this journal
- Development and Multinational Validation of an Ensemble Deep Learning Algorithm for Detecting and Predicting Structural Heart Disease Using Noisy Single-lead Electrocardiograms 97%
- International Evaluation Of An Artificial Intelligence-Powered Ecg Model Detecting Occlusion Myocardial Infarction 97%
- Simple Models Versus Deep Learning in Detecting Low Ejection Fraction From The Electrocardiogram 96%
Similar papers in this journal
- A genotype-phenotype taxonomy of hypertrophic cardiomyopathy 93%
- A Polygenic Risk Score for Coronary Artery Disease Improves the Prediction of Early-Onset Myocardial Infarction and Mortality in Men 93%
- Machine Learning to Understand Genetic and Clinical Factors Associated with the Pulse Waveform Dicrotic Notch 93%
Similar papers in this journal
- Cohort Design and Natural Language Processing to Reduce Bias in Electronic Health Records Research: The Community Care Cohort Project 95%
- Multicenter Evaluation of Interpretable AI for Coronary Artery Disease Diagnosis from PET Biomarkers 94%
- Identification of Digital Twins to Guide Interpretable AI for Diagnosis and Prognosis in Heart Failure 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.