Intra-Aortic Pressure Forecasting Using Back-EMF Pump Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack the ability to accurately predict intra-aortic pressure in patients receiving hemodynamic support from a transvalvular micro-axial heart pump, which is crucial for early detection of conditions like acute decompensated heart failure and optimizing treatment strategies.
Innovation Solution
A system utilizing a transvalvular micro-axial heart pump with a pressure sensor and machine learning algorithms, such as deep learning models, to predict intra-aortic pressure based on pressure and motor speed measurements, enabling real-time adjustments to motor speed settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning algorithms are applied to predict intra-aortic pressure, then forecasting accuracy is improved, but device complexity increases
Solution Approach 1:
A trained machine learning model serves as an intermediary between raw sensor measurements (pressure and motor speed) and the prediction output. The model processes the complex non-linear relationships in the time series data, enabling accurate short-term forecasting without requiring direct complex computational analysis of the raw signals.
Solution Approach 2:
The machine learning model is trained in advance on historical intra-aortic pressure and motor speed data before deployment. This preliminary training phase allows the system to learn complex patterns and relationships, so that during actual operation, predictions can be generated rapidly with minimal real-time computational complexity.
2Measurement precision
If high frequency intra-aortic blood pressure time series is used for prediction, then measurement precision is improved, but data noise and non-stationarity increase
Solution Approach 1:
The system transforms the raw high-frequency pressure signal into a derived motor speed time series using back electromotive force (EMF) measurements. This parameter transformation converts a noisy, non-stationary pressure signal into a more stable derivative signal that captures the same physiological information with reduced noise and improved stationarity for modeling purposes.
Solution Approach 2:
The system replaces direct analysis of the mechanical pressure signal with an electrical measurement approach. By measuring the back EMF of the motor, the system derives motor speed information that correlates with pressure changes but is obtained through an electrical sensing mechanism that is less susceptible to the noise and artifacts affecting direct pressure measurements.
3Measurement precision
If deep learning models are used for short-term prediction, then forecasting accuracy is improved, but computational energy consumption increases
Solution Approach 1:
The deep learning model is trained offline on substantial datasets before deployment in the implantable device. This preliminary training transfers the computationally intensive learning process to an external system, allowing the implanted device to perform only inference operations with minimal energy consumption while maintaining high prediction accuracy.
Solution Approach 2:
The system uses a simplified version or pre-trained model that captures the essential predictive capabilities without requiring the full computational power of the training process. The model structure is optimized for efficient inference on resource-constrained implantable devices while preserving the accuracy benefits of deep learning.
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
Enhances the ability of clinicians to forecast patient conditions, allowing for early intervention and improved patient management by predicting intra-aortic pressure with short-term accuracy.
Implementation Method 1
obtain a set of intra-aortic pressure measurements corresponding to pressure values measured by the pressure sensor
Implementation Method 2
obtain a set of motor speed measurements corresponding to rotational speeds of the motor
Implementation Method 3
predict, using a trained machine learning model, an intra-aortic pressure of a patient based on the sets of intra-aortic pressure and motor speed measurements
Data Source
AI summary
Aspects of the present disclosure describe systems and methods for predicting an intra-aortic pressure of a patient receiving hemodynamic support from a transvalvular micro-axial heart pump. In some implementations, an intra-aortic pressure time series is derived from measurements of a pressure sensor of the transvalvular micro-axial heart pump and a motor speed time series is derived from a measured back electromotive force of a motor of the transvalvular micro-axial heart pump. Furthermore, in some implementations, machine learning algorithms, such as deep learning, are applied to the intra-aortic pressure and motor speed time series to accurately predict an intra-aortic pressure of the patient. In some implementations, the prediction is short-term (e.g., approximately 5 minutes in advance).