Non-invasive Cardiac Health Assessment via ML-Enhanced PPG Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current invasive intracardiac pressure monitoring technologies are limited in their ability to accurately assess left-sided filling pressures, particularly in cases of increased pulmonary resistance, and are costly, invasive, and unable to identify precipitating factors of heart failure hospitalizations, necessitating a non-invasive and affordable solution for real-time cardiac health assessment.
Innovation Solution
A non-invasive cardiac health assessment system utilizing a combination of sensor technologies and machine learning algorithms to train a device based on intracardiac pressure data, incorporating invasive and non-invasive measurements to estimate cardiac health without invasive interventions, using sensors like Photoplethysmography (PPG), soundwave transducers, and Inertial Measurement Units (IMU) to capture cardiac health data and adjust medication accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If invasive intracardiac pressure monitoring is used, then measurement precision of intracardiac pressure is improved, but device complexity and patient discomfort increase
Solution Approach 1:
The patent uses an intermediary machine learning model that translates non-invasive PPG sensor data into estimates of intracardiac pressure. This intermediary approach allows the system to achieve invasive-level measurement precision without actually using invasive sensors, thereby resolving the contradiction between measurement accuracy and device complexity.
Solution Approach 2:
The patent replaces the mechanical invasive pressure sensing system with an optical PPG-based system combined with machine learning algorithms. This substitution eliminates the need for physical intrusion into the cardiovascular system while maintaining the ability to estimate intracardiac pressure, thus resolving the contradiction between measurement precision and device complexity.
2Reliability
If invasive pressure monitoring is used, then reliability of cardiac health assessment is improved, but ease of operation deteriorates
Solution Approach 1:
The system uses a smartphone application that patients can operate independently to perform cardiac assessments. The automated machine learning model processes the data without requiring medical professional intervention, making the reliable cardiac health assessment accessible to patients themselves and thus improving ease of operation.
3Measurement precision
If invasive monitoring technologies are deployed, then diagnostic accuracy is improved, but cost efficiency worsens
Solution Approach 1:
The patent employs inexpensive, disposable PPG sensors integrated into smartphone applications rather than expensive, reusable invasive pressure monitoring equipment. This approach maintains diagnostic accuracy through sophisticated machine learning while dramatically reducing the cost per assessment, thus resolving the contradiction between diagnostic accuracy and cost efficiency.
4Ease of operation
If non-invasive sensors are used, then ease of operation is improved, but measurement precision of intracardiac pressure deteriorates
Solution Approach 1:
The patent transforms the relationship between non-invasive PPG parameters and intracardiac pressure through machine learning model training. By learning complex non-linear relationships from training data, the system enables accurate intracardiac pressure estimation from simple non-invasive optical measurements, thus resolving the contradiction between ease of operation and measurement precision.
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 system provides accurate, real-time cardiac health monitoring, reducing hospitalizations and improving diagnostic and treatment capabilities for heart failure patients by leveraging machine learning to interpret non-invasive data similarly to invasive readings, thus addressing the limitations of existing technologies.
Implementation Method 1
a wearable device including a Photoplethysmography (PPG) sensor to measure volumetric variations of blood circulation
Implementation Method 2
a soundwave transducer configured to capture auscultation signals indicative of cardiac health of the user
Implementation Method 3
an Inertial Measurement Unit (IMU) sensor configured to capture seismic auscultation signals indicative of the cardiac health of the user
Data Source
AI summary
The present disclosure relates to cardiac health assessment system for use with a handheld electronic device for assessing cardiac health of a user and a method for assessing cardiac health of a user. The disclosure further relates to systems and methods for training a machine learning model to estimate intracardiac pressure data.