PPG Signal Processing for Cardiac Dysfunction Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting left ventricular systolic dysfunction (LVSD) are invasive, inaccurate, and often go undiagnosed, with existing tests having a low area under the receiver operating characteristic curve (AUC) of 0.6-0.7, leading to missed diagnoses and increased mortality.
Innovation Solution
A computer-implemented method using machine-learned models that process photoplethysmogram (PPG) signals, potentially combined with electrocardiogram (ECG) and demographic data, to predict cardiac dysfunction, such as LVSD, in wearable devices, providing a more accurate prediction with an AUC of approximately 0.86 and surfacing notifications for early detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine-learned models process PPG signals to generate cardiac dysfunction predictions, then measurement precision of cardiac dysfunction detection is improved, but device complexity increases
Solution Approach 1:
The patent introduces machine-learned models as an intermediary between the simple PPG signal acquisition and the complex task of cardiac dysfunction detection. The model acts as a mediator that transforms raw PPG signals into accurate ejection fraction predictions, thereby improving measurement precision without requiring complex invasive hardware
Solution Approach 2:
The patent replaces complex mechanical/invasive detection systems with an optical-based PPG sensing system combined with computational processing. Instead of using invasive pressure sensors or complex imaging equipment, the system uses light absorption measurements processed through machine learning to achieve accurate cardiac dysfunction detection
2Ease of operation
If PPG signals are used for cardiac dysfunction detection, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent transforms the PPG signal parameters through machine learning processing to extract meaningful cardiac dysfunction indicators. By changing the parameter representation from raw optical absorption values to predicted ejection fraction and cardiac rhythm metrics, the system maintains ease of operation while achieving high measurement precision
Solution Approach 2:
The patent creates a computational model that copies the relationship between PPG signals and cardiac dysfunction based on training data. This virtual copy allows the simple PPG sensor to provide accurate cardiac dysfunction detection by referencing the learned patterns from extensive training datasets
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method significantly improves the accuracy of LVSD detection, reduces false positives and negatives, and enables early intervention, decreasing mortality and the need for invasive testing, while being accessible through widespread wearable technology.
Implementation Method 1
obtaining, from a sensor, photoplethysmogram (PPG) signals indicative of a cardiac rhythm of the user
Data Source
AI summary
A computer-implemented method for detecting a cardiac dysfunction in a user includes obtaining, from a sensor, photoplethysmogram (PPG) signals indicative of a cardiac rhythm of the user. The computer-implemented method further includes processing, in a computing device, the PPG signals to generate a cardiac dysfunction prediction for the user based, at least in part, on the PPG signals. The computer-implemented method further includes providing, via an annunciator, the cardiac dysfunction prediction for the user as an output.