ECG Data Compression via Probability Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ECG monitoring systems are inadequate for long-term, continuous recording of cardiac activity, particularly for capturing sporadic cardiac events, due to discomfort, inefficiency, and limited ability to sense low-amplitude P-waves, which are crucial for accurate diagnosis of cardiac rhythm disorders.
Innovation Solution
A lightweight wearable monitor with a flexible extended wear electrode patch and a reusable recorder that can be easily positioned on the sternum, optimized to sense P-waves and QRS intervals, and equipped with a two-step data compression algorithm for efficient storage and transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If conventional ECG monitoring systems are used for long-term continuous recording, then storage requirements and power consumption increase significantly, but the ability to capture sporadic cardiac events and maintain diagnostic accuracy deteriorates
Solution Approach 1:
The patent extracts and removes redundant information from ECG signals by identifying and eliminating duplicate or highly similar waveform segments. Only unique or diagnostically significant portions of the ECG data are retained for storage, dramatically reducing the quantity of data that must be stored while preserving all essential diagnostic information for capturing sporadic cardiac events.
Solution Approach 2:
The system discards redundant ECG data segments that have been previously recorded and are highly similar to existing stored data. By comparing incoming ECG waveforms against stored reference patterns, the system identifies and discards duplicate segments, recovering storage space for potentially unique diagnostic events while maintaining the ability to reconstruct complete ECG records when needed.
2Measurement precision
If high-resolution ECG data is recorded continuously to maintain diagnostic accuracy, then storage capacity and power consumption increase, but the system becomes less practical for extended wear
Solution Approach 1:
The patent implements periodic compression and filtering operations at strategically determined intervals rather than continuously processing every ECG sample. The system periodically analyzes ECG segments for redundancy and applies compression algorithms only when diagnostic information permits, reducing overall power consumption while maintaining measurement precision for all diagnostically relevant cardiac events.
Solution Approach 2:
The system dynamically changes processing parameters based on the detected ECG signal characteristics and current storage conditions. When storage is abundant and power is available, higher resolution recording is maintained. When storage becomes constrained or power consumption needs reduction, the system adjusts compression ratios and filtering parameters to balance diagnostic accuracy with energy efficiency and extended wear practicality.
3Reliability
If complete ECG data is stored without compression to maintain diagnostic accuracy, then storage space is rapidly consumed, but the monitoring system becomes cumbersome and less user-friendly
Solution Approach 1:
The patent creates compressed representative copies of ECG data that retain all essential diagnostic features while occupying minimal storage space. By identifying characteristic waveform patterns and storing only unique or abnormal segments with full fidelity while compressing normal segments, the system maintains diagnostic accuracy for detecting sporadic cardiac events while dramatically reducing storage requirements, making the system practical and user-friendly for extended wear.
Data Source
AI summary
A monitor recorder-implemented method for electrocardiography data compression and an electrocardiography monitor recorder with integral data compression are provided. A series of data items are obtained, each of the data items associated with a magnitude of an ECG signal sensed by a monitor recorder. A range is set for an initial one of the data items in the series. Each of the data items remaining in the series is processed, including: obtaining an estimation of probabilities of the data items appearing next to that data item in the series; dividing the further range into sub-ranges, each sub-range representing a fraction of the further range proportional to the probabilities of the next data items; selecting the sub-range corresponding to the data item next to that data item in the series; and representing the next data item by the selected sub-range in a non-volatile memory.