# Wearable ECG-PPG Deep Learning Model for Cardiac Index-Based Noninvasive Cardiac Output Estimation in Cardiac Surgery Patients

## TL;DR

A wearable device using ECG and PPG signals with deep learning can estimate cardiac output noninvasively, with better accuracy when using cardiac index normalization.

## Contribution

A novel wearable ECG–PPG fusion model with deep learning for noninvasive cardiac output estimation, showing improved agreement with invasive measurements through CI normalization.

## Key findings

- CI prediction model showed the best performance in estimating cardiac index.
- Indirect CO estimation using CI normalization met the predefined error benchmark (PE < 30%).

## Abstract

What are the main findings?

A lightweight wearable ECG–PPG fusion model estimated cardiac index (CI) and cardiac output (CO), with the best performance observed for CI prediction.

CI-based normalization improved agreement with thermodilution reference measurements, and indirect CO reconstruction met the predefined benchmark (percentage error, PE < 30%).

What are the implications of the main findings?

Wearable ECG–PPG monitoring combined with deep learning may enable catheter-free, continuous trending of CO/CI in controlled perioperative settings.

CI-normalized modeling may improve generalizability and support future noninvasive hemodynamic monitoring tools, pending external and ambulatory validation.

Accurate cardiac output (CO) measurement is vital for hemodynamic management; however, it usually requires invasive monitoring, which limits its continuous and out-of-hospital use. Wearable sensors integrated with deep learning offer a noninvasive alternative. This study developed and validated a lightweight deep learning model using wearable electrocardiography (ECG) and photoplethysmography (PPG) signals to predict CO and examined whether cardiac index-based normalization (Cardiac Index (CI) = CO/body surface area) improves performance. Twenty-seven patients who underwent cardiac surgery and had pulmonary artery catheters were prospectively enrolled. Single-lead ECG (HiCardi+ chest patch) and finger PPG (WristOx2 3150) were recorded simultaneously and processed through an ECG–PPG fusion network with cross-modal interaction. Three models were trained as follows: (1) CI prediction, (2) direct CO prediction, and (3) indirect CO prediction. The total number of CO = predicted CI × body surface area. Reference values were derived from thermodilution. The CI model achieved the best performance, and the indirect CO model showed significant reductions in error/agreement metrics (MAE/RMSE/bias; p < 0.0001), while correlation-based metrics are reported descriptively without implying statistical significance. The Pearson correlation coefficient (PCC) and percentage error (PE) for the indirect CO estimates (PCC = 0.904; PE = 23.75%). The indirect CO estimates met the predefined PE < 30% agreement benchmark for method-comparison; this is not a universal clinical standard. These results demonstrate that wearable ECG–PPG fusion deep learning can achieve accurate, noninvasive CO estimation and that CI-based normalization enhances model agreement with pulmonary artery catheter measurements, supporting continuous catheter-free hemodynamic monitoring.