Classifier to predict cardiac output through photoplethysmography waveform analysis

Joshua H. Chiu; Kimberly L. Branan; Chin-To Hsiao; Gerard L. Coté

doi:10.1117/12.3001861

27 March 2024 Classifier to predict cardiac output through photoplethysmography waveform analysis

Joshua H. Chiu, Kimberly L. Branan, Chin-To Hsiao, Gerard L. Coté

Proceedings Volume 12850, Optical Diagnostics and Sensing XXIV: Toward Point-of-Care Diagnostics; 128500J (2024) https://doi.org/10.1117/12.3001861
Event: SPIE BiOS, 2024, San Francisco, California, United States

Conference Poster

Abstract

In this study, features were extracted from PPG waveforms and used to train different classifier models, where cardiac output (CO) was classified into three ranges: low, healthy, and high. We used a curated dataset from the PhysioNet platform, Medical Information Mart for Intensive Care III (MIMIC-III) Matched database, which contains PPG waveforms and CO measurements. Specifically, there were 16 viable patients with over 184 hours of synchronous PPG and CO measurements. The data was then categorized into three distinct classes: Low CO (< 5 L/min), Healthy CO (5-6 L/min), and High CO (> 6 L/min). MATLAB’s Classification Learner application was used to implement and compare different classification techniques where thirty pre-configured models were compared, including SVMs, KNNs, Ensemble, Linear and Logistical Regression, and Neural Networks. From all the tested models, a Bagged Trees Ensemble model was determined to have the highest validation and testing accuracy (87.7% and 88.2%, respectively). It is noteworthy that, with our approach, there was no calibration needed and since the validation and testing accuracies were similar, this suggests that the selected model did not overfit the data.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Joshua H. Chiu, Kimberly L. Branan, Chin-To Hsiao, and Gerard L. Coté "Classifier to predict cardiac output through photoplethysmography waveform analysis", Proc. SPIE 12850, Optical Diagnostics and Sensing XXIV: Toward Point-of-Care Diagnostics, 128500J (27 March 2024); https://doi.org/10.1117/12.3001861

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available