Annotated PPG Signal Generation via ECG Beat Pairing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for annotating photoplethysmography (PPG) signals are unreliable and inconsistent, limiting their use in medical diagnostics, particularly for detecting cardiac rhythm disorders beyond atrial fibrillation, due to sensitivity to recording conditions and lack of reliable training data, with existing solutions relying on mapping ECG annotations onto PPG signals without addressing noise and artefacts.
Innovation Solution
A method for generating annotated PPG signals by synchronizing ECG and PPG recordings, detecting cardiac beats, pairing them, and deriving annotations based on the nature of ECG segment annotations, while considering the quality and accuracy of the pairing, to produce more reliable and consistent annotations for training neural networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If ECG annotations are mapped onto PPG signals, then annotation availability is improved, but annotation reliability deteriorates due to noise and artefacts in PPG signals
Solution Approach 1:
The patent introduces an intermediary system that processes both ECG and PPG signals through beat detection and pairing mechanisms. This intermediary layer transforms direct ECG-PPG annotation mapping into a controlled process where only reliable signal portions are annotated, using the ECG signal as a reference to guide PPG annotation while filtering out noisy segments through quality assessment
Solution Approach 2:
The patent implements feedback through a quality assessment mechanism that evaluates PPG signal portions and determines whether they are suitable for annotation. The system uses pairing success between ECG and PPG beats as feedback to decide whether to apply ECG annotations, creating a closed-loop process that improves annotation reliability by excluding low-quality signal segments
2Productivity
If machine learning classifiers are trained with available PPG data, then classification capability is improved, but classification accuracy deteriorates due to inconsistent and unreliable annotations
Solution Approach 1:
The patent applies preliminary action by pre-processing PPG signals through quality assessment and selective annotation before training machine learning classifiers. The system identifies and annotates only high-quality PPG signal portions using the reliable ECG-PPG pairing method, ensuring that training data is prepared in advance with verified accuracy, which directly improves classification performance
Solution Approach 2:
The patent changes the parameter of data quality through selective annotation based on signal pairing success. By transforming the training dataset to include only portions where ECG-PPG beat pairing was successful, the system alters the quality parameters of the training data, thereby improving classification accuracy without sacrificing productivity
3Reliability
If contact PPG is used for medical diagnosis, then measurement reliability is improved, but susceptibility to artefacts worsens due to motion and positioning issues
Solution Approach 1:
The patent extracts and removes artefact-contaminated portions of the PPG signal through quality assessment. By identifying signal segments where beat pairing fails or quality metrics indicate contamination, the system extracts only the clean, reliable portions for annotation and training, effectively eliminating the harmful effect of motion and positioning artefacts
Solution Approach 2:
The patent converts the presence of artefacts into a benefit by using them as indicators for quality assessment. The failure of beat pairing or presence of artefacts automatically identifies problematic signal portions, which then serve as negative training examples or are excluded from training, transforming the harmful artefacts into a mechanism for improving overall system robustness
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
This approach results in more reliable and consistent annotated PPG signals that can accurately detect various heart rhythms and clinical disorders, improving medical diagnosis by providing better training data for machine learning tools and identifying insufficient quality portions in PPG signals.
Implementation Method 1
PPG makes use of light absorption by blood to track these volumetric changes
Data Source
AI summary
A computer-implemented method for generating an annotated photoplethysmography signal, includes: recording an electrocardiogram or ECG signal; recording a photoplethysmography or PPG signal, semi-synchronously with the recording of the ECG signal; annotating segments in the ECG signal either algorithm-based or expert-based; time-aligning the PPG signal and the ECG signal; detecting ECG beats in the ECG signal; detecting PPG beats in the PPG signal; pairing the ECG beats onto the PPG beats; deriving annotations for PPG signal segments based on the nature of the ECG segment annotations and on how the ECG beats can be paired with the PPG beats; and annotating the PPG signal segments using the annotations, thereby generating the annotated PPG signal.