Reconstructing Surface ECG from Endocardial Far Field Signals

The patent applies preliminary action by pre-calculating and storing transfer matrices during a learning phase that captures propagation time delays and lead position variations. These pre-computed matrices are then reused during actual ECG reconstruction, avoiding real-time complex calculations while maintaining high accuracy. The learning phase stores the relationship between EGM and ECG signals in different conditions, enabling accurate reconstruction without real-time computational burden.

The patent changes parameters by using adaptive transfer matrices that are updated based on detected lead positions and patient-specific characteristics. The system adjusts the reconstruction parameters (transfer matrices) according to varying conditions such as lead displacement or changes in body geometry, thereby maintaining measurement precision without requiring complex real-time computation. This allows the system to adapt to different physiological states while using computationally efficient methods.

The patent pre-computes transfer matrices during a learning phase that captures propagation time delays and lead position variations. These pre-calculated matrices are stored and reused during actual ECG reconstruction, eliminating the need for complex real-time calculations. The system performs the computationally intensive work in advance, when accuracy is critical but computational resources can be dedicated to learning rather than real-time processing.

The patent creates simplified copies of the complex physiological propagation processes by using pre-computed transfer matrices that model the relationship between EGM and ECG signals. Instead of simulating the actual complex electrical propagation through tissues in real-time, the system uses copied relationships from the learning phase, which capture the essential characteristics without requiring complex computational models during operation.

The patent extracts the ECG signal acquisition function from the complex process of surface electrode placement and Holter recorder usage. By utilizing the already-implanted device's electrodes and processing capabilities, the system eliminates the need for external ECG recording equipment and preparation procedures. The implantable device itself performs ECG reconstruction from its existing EGM signals, removing the separate ECG recording step entirely.

The patent makes the implantable device universal by enabling it to perform multiple functions: its original pacing/defibrillation function plus ECG reconstruction and monitoring. The same implanted electrodes and processing unit that were designed for cardiac therapy are now also used for ECG signal acquisition and reconstruction, eliminating the need for separate ECG recording equipment and simplifying the overall system.

The patent enables the implantable device to perform ECG reconstruction autonomously using its own embedded processing capabilities. The device automatically reconstructs ECG signals from its EGM measurements without requiring external equipment or practitioner intervention for signal acquisition. This self-service capability allows continuous or frequent monitoring without scheduling constraints, eliminating the time loss associated with arranging and performing scheduled ECG recordings.

The patent extracts the ECG recording function from the time-consuming process of scheduled clinic visits with surface electrode placement. By using the implantable device's existing sensors and processing unit, the system performs ECG reconstruction at the patient's home or in routine device checks, removing the need for dedicated ECG recording appointments and significantly reducing the time intervals between effective monitoring opportunities.