Cardiac Function Assessment via Spatiotemporal Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging techniques, particularly 2-dimensional echocardiography, are burdensome for analyzing cardiac health and do not effectively capture changes over time, such as those in a beating heart, leading to inconsistent and error-prone diagnoses.
Innovation Solution
A cardiovascular analysis application that receives patient echocardiograms as video segments, extracts spatiotemporal features, and uses trained classification models to assess ejection fraction and hypertrophic cardiomyopathy (HCM), providing indications of insufficient ejection fraction and HCM presence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If 2-dimensional echocardiography scans are used for cardiac analysis, then imaging can be obtained, but the analysis becomes burdensome and cannot capture changes over time
Solution Approach 1:
The patent transitions from static 2D images to dynamic video segments that capture the beating heart over time. The system processes video segments containing multiple heartbeat cycles, enabling detection of temporal changes in cardiac function while maintaining diagnostic accuracy through automated spatiotemporal feature extraction.
2Measurement precision
If expert interpretation is used for echocardiogram analysis, then accurate diagnosis can be achieved, but the process is time-consuming and resource-intensive
Solution Approach 1:
The system performs self-service through automated classification models that independently analyze echocardiogram video segments without requiring expert intervention. The trained models extract spatiotemporal features and classify cardiac conditions automatically, maintaining high detection accuracy while significantly reducing analysis time and resource requirements.
3Reliability
If video segments with multiple heartbeat cycles are analyzed, then temporal changes can be captured, but the amount of data to process increases
Solution Approach 1:
The patent extracts and processes only the essential spatiotemporal features from video segments containing multiple heartbeat cycles. By focusing on characteristic motion patterns and temporal variations rather than processing every pixel of raw video data, the system captures functional changes accurately while minimizing data processing requirements.
Data Source
AI summary
A cardiovascular analysis application receives patient echocardiograms as video segments and performs an analysis for cardiovascular health based on factors related to an ejection fraction and hypertrophic cardiomyopathy based on the images in the video segment and spatiotemporal features extracted from the images through several heartbeat cycles on the video segments. The models are trained on a corpus of previous echocardiograms including labels indicative of the ejection fraction and physiological markers associated with HCM, such as the cardiac wall thickness and clarity. Based on a correspondence with the model, a result is rendered indicative of whether the patient video segment has an insufficient ejection fraction and whether a presence of HCM is exhibited.