Intra-aortic pressure forecasting
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).
1. A system comprising one or more processors configured to:
receive data from a transvalvular micro-axial heart pump during a period of time when the transvalvular micro-axial heart pump is at least partially located in a heart of a patient;
derive a set of features from the received data, wherein the set of features comprises (a) pressure measurements corresponding to pressure values measured by a pressure sensor of the transvalvular micro-axial heart pump, (b) motor speed measurements corresponding to rotational speeds of a motor of the transvalvular micro-axial heart pump, or (c) motor current measurements corresponding to an energy intake of the motor; and
predict, using a trained machine learning model, a cardiac condition of the patient based on the derived set of features, wherein the machine learning model is trained on a data set comprising increasing sequences, decreasing sequences, and stationary sequences, and wherein each sequence comprises pressure measurements, motor speed measurements, or motor current measurements.
2. The system of claim 1 , wherein the transvalvular micro-axial heart pump comprises a tube, an inlet area having one or more openings through which blood may be drawn into the tube by the motor, and an outlet area having one or more openings through which blood may be expelled from the tube by the motor.
3. The system of claim 1 , wherein the machine learning model is a deep learning model.
4. The system of claim 3 , wherein the deep learning model is an Autoregressive Integrated Moving Average (ARIMA) model, a Deep Neural Network (DNN) model, a Recurrent Sequence to Sequence model, a Recurrent Sequence to Sequence model with Attention, a Transformer model, a Temporal Convolutional Neural Network (TCN) model, or a Convolutional Neural Pyramid model.
5. The system of claim 1 , wherein a sequence is increasing if the measurements within that sequence increase by more than a first predetermined threshold, wherein a sequence is decreasing if the measurements within that sequence decrease by more than a second predetermined threshold, and wherein a sequence is stationary if the measurements within that sequence do not (a) increase by more than the first predetermined threshold or (b) decrease by more than the second predetermined threshold.
6. The system of claim 1 , further comprising:
a display configured to display the predicted cardiac condition of the patient.
7. The system of claim 6 , wherein the display is further configured to display an alert when the cardiac condition of the patient is predicted to deteriorate.
8. The system of claim 1 , wherein the one or more processors are further configured to automatically adjust a speed setting of the motor based on the predicted cardiac condition of the patient.
9. The system of claim 1 , wherein the predicted cardiac condition of the patient is a prediction of a likelihood of the patient experiencing an acute hypotensive episode (AHE) or acute decompensated heart failure (ADHF).
10. A system comprising one or more processors configured to:
receive data from a transvalvular micro-axial heart pump during a period of time when the transvalvular micro-axial heart pump is at least partially located in a heart of a patient;
derive a set of features from the received data, wherein the set of features comprises (a) pressure measurements corresponding to pressure values measured by a pressure sensor of the transvalvular micro-axial heart pump, (b) motor speed measurements corresponding to rotational speeds of a motor of the transvalvular micro-axial heart pump, or (c) motor current measurements corresponding to an energy intake of the motor; and
predict, using a trained machine learning model, a cardiac condition of the patient based on the derived set of features, wherein the machine learning model is a Recurrent Sequence to Sequence model with a Legendre Memory Unit (LMU).
11. The system of claim 10 , wherein the one or more processors are further configured to automatically adjust a speed setting of the motor based on the predicted cardiac condition of the patient.
12. A system comprising one or more processors configured to:
receive data from a transvalvular micro-axial heart pump during a period of time when the transvalvular micro-axial heart pump is at least partially located in a heart of a patient;
derive a set of features from the received data, wherein the set of features comprises a) pressure measurements corresponding to pressure values measured by a pressure sensor of the transvalvular micro-axial heart pump, (b) motor speed measurements corresponding to rotational speeds of a motor of the transvalvular micro-axial heart pump, or (c) motor current measurements corresponding to an energy intake of the motor; and
predict, using a trained machine learning model, a cardiac condition of the patient based on the derived set of features, wherein the machine learning model is trained on a data set comprising only increasing and decreasing sequences, and wherein each sequence comprises pressure measurements, motor speed measurements, or motor current measurements.
13. The system of claim 12 , wherein a sequence is increasing if the measurements within that sequence increase by more than a first predetermined threshold, and wherein a sequence is decreasing if the measurements within that sequence decrease by more than a second predetermined threshold.
14. The system of claim 12 , wherein the one or more processors are further configured to automatically adjust a speed setting of the motor based on the predicted cardiac condition of the patient.
15. A method comprising:
receiving data from a transvalvular micro-axial heart pump during a period of time when the transvalvular micro-axial heart pump is at least partially located in a heart of a patient;
deriving a set of features from the received data, wherein the set of features comprises (a) pressure measurements corresponding to pressure values measured by a pressure sensor of the transvalvular micro-axial heart pump, (b) motor speed measurements corresponding to rotational speeds of a motor of the transvalvular micro-axial heart pump, or (c) motor current measurements corresponding to an energy intake of the motor; and
predicting, using a trained machine learning model, a cardiac condition of the patient based on the derived set of features, wherein the machine learning model is trained on a data set comprising increasing sequences, decreasing sequences, and stationary sequences, and wherein each sequence comprises pressure measurements, motor speed measurements, or motor current measurements.
16. The method of claim 15 , wherein the transvalvular micro-axial heart pump comprises a tube, an inlet area having one or more openings through which blood may be drawn into the tube by the motor, and an outlet area having one or more openings through which blood may be expelled from the tube by the motor.
17. The method of claim 15 , further comprising:
adjusting an amount of a medication provided to the patient based on the predicted cardiac condition of the patient.
18. The method of claim 15 , further comprising:
decreasing a speed of the motor if the cardiac condition of the patient is predicted to improve.
19. The method of claim 15 , further comprising:
increasing a speed of the motor if the cardiac condition of the patient is predicted to deteriorate.
20. A non-transitory computer readable storage medium having instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform the method of claim 15 .