Journal of cardiology · 2025
Novel application of contactless remote photoplethysmography in identifying heart failure with left ventricular systolic dysfunction.
СтатьяПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Chang Ting-Yung, Wu Yi-Chiao, Chung Meng-Liang, Wu Bing-Fei, Cheng Hao-Min, Lee Wui-Chiang, Chang Shih-Lin
Аннотация
BACKGROUND: Left ventricular ejection fraction (LVEF) serves as a prognostic marker in heart failure (HF) patients, and screening for undiagnosed HF and meticulous monitoring are crucial for improving outcomes in HF patients. The novel contactless camera-based photoplethysmography (PPG), measured by FHVitals SDK (FaceHeart Corp., Taipei, Taiwan), is the first remote PPG (rPPG) method approved by the U.S. Food and Drug Administration (FDA) for measuring heart rate and respiratory rate. A clinical study also demonstrated the accuracy in detecting atrial fibrillation. The aim of this study was to evaluate the accuracy of rPPG in identification of HF with LVEF <50 %.
METHODS: Sixty-eight participants were enrolled and divided into 2 groups (HF group with LVEF: <50 % and control group). All facial image data were processed by FHVitals SDK for pulsatile signal extraction. We employed support vector machine (SVM), decision tree, and random forest algorithms as classifiers for the HF prediction task.
RESULTS: The study cohorts contained 68 participants, and 5 devices recorded each participant, which included a total of 340 video samples. Compared to other classifiers, the overall metrics such as area under the curve (AUC) or f1-score were both better with SVM (AUC of 0.9, accuracy of 89.71 %, and f1-score of 0.9). There was no obvious correlation between misclassification and underlying diseases or medical treatment.
CONCLUSION: This is the first study to demonstrate that the FDA-approved contactless rPPG could be used to identify patients with recent HF with LV systolic dysfunction. This technology could not only assist in screening HF patients, but be helpful in remote monitoring at home.
METHODS: Sixty-eight participants were enrolled and divided into 2 groups (HF group with LVEF: <50 % and control group). All facial image data were processed by FHVitals SDK for pulsatile signal extraction. We employed support vector machine (SVM), decision tree, and random forest algorithms as classifiers for the HF prediction task.
RESULTS: The study cohorts contained 68 participants, and 5 devices recorded each participant, which included a total of 340 video samples. Compared to other classifiers, the overall metrics such as area under the curve (AUC) or f1-score were both better with SVM (AUC of 0.9, accuracy of 89.71 %, and f1-score of 0.9). There was no obvious correlation between misclassification and underlying diseases or medical treatment.
CONCLUSION: This is the first study to demonstrate that the FDA-approved contactless rPPG could be used to identify patients with recent HF with LV systolic dysfunction. This technology could not only assist in screening HF patients, but be helpful in remote monitoring at home.