System, method, and computer software code for predicting an acute hypotensive episode
A system for determining onset of an acute hypotensive episode, the system includes a plurality of collectors configured to determine heart rate, mean arterial pressure, stroke volume, total peripheral resistance, and/or age of a patient, an estimate model configured to receive data from the collectors, a simulate model configured to receive information from the estimate model to simulate operation of an effector portion and a neural portion of an arterial baroreceptor reflex pathway, and a prediction model configured predict onset of an acute hypotensive episode based on results from the simulate model. A method and a computer software code for determining onset of an acute hypotensive episode are also disclosed.
1. A method for determining onset of an acute hypotensive episode, the method comprising:
determining heart rate and mean arterial pressure of a patient at a given time using one or more sensors coupled to the patient;
estimating, using at least one processor, an onset of an acute hypotensive episode based on the measured heart rate and mean arterial pressure at the given time in association with heart rate and mean arterial pressure time-series data prior to the given time within a predetermined period;
simulating, using the processor, an evolution of mean arterial pressure based on a mathematical model of a neural portion of an arterial baroreceptor reflex pathway for a predetermined prediction period and further based on a number of repetitive runs, wherein the neural portion parameters are random coefficients of a linear system described as an nth order differential equation with random coefficients calculated via a maximum likelihood estimator using past read heart rate and mean arterial pressure of the patient over a predetermined estimation period;
predicting, using the processor, the acute hypotensive episode based on results from simulating the onset of the acute hypotensive episode; and
reporting, using the processor, the predicted acute hypotensive episode.
2. The method according to claim 1 , wherein the predetermined prediction period is at least a minimum length of the acute hypotensive episode defined as thirty minutes.
3. The method according to claim 1 , wherein the predetermined estimation period is at most a length of a recorded heart rate and mean arterial pressure data since an initiation of vital sign recording for the patient.
4. The method according to claim 1 , wherein an order of the differential equations is greater than or equal to 1 and less than or equal to a pre-determined estimation window.
5. The method according to claim 1 , further comprising prohibiting predicting the acute hypotensive episode when the measured mean arterial pressure is below a predetermined level such as 60 mmHg for acute hypotensive episode (AHE) for a predetermined period such as thirty minutes for AHE.
6. The method according to claim 1 , wherein predicting the acute hypotensive episode further comprises simulating an evolution of mean arterial pressure using a mathematical model of an effector portion of an arterial baroreceptor reflex pathway.
7. The method according to claim 1 , wherein predicting the acute hypotensive episode further comprises simulating an evolution of mean arterial pressure using a Gaussian random generator and constant total peripheral resistance.
8. The method according to claim 1 , wherein predicting the acute hypotensive episode further comprising determining whether the acute hypotensive episode has occurred within a prediction window.
9. The method according to claim 1 , further comprises filtering data associated with determining heart rate and mean arterial pressure.
10. A non-transitory computer readable medium having computer software code embodied thereon that is configured for execution with a processor designated for collecting and controlling sharing of data associated with a patient's medical condition, the computer software code comprising instructions for performing a method comprising:
determining heart rate and mean arterial pressure of a patient at a given time using data obtained from one or more sensors;
estimating an onset of an acute hypotensive episode based on the measured heart rate and mean arterial pressure at the given time in association with heart rate and mean arterial pressure time-series data prior to the given time within a predetermined period;
simulating an evolution of the measured mean arterial pressure based on a mathematical model of a neural portion of an arterial baroreceptor reflex pathway for a predetermined prediction period and further based on a number of repetitive runs, wherein the neural portion parameters are random coefficients of a linear system described as an nth order differential equation with random coefficients calculated via a maximum likelihood estimator using past read heart rate and mean arterial pressure of the patient over a predetermined estimation period; and
predicting the acute hypotensive episode based on results from simulating the onset of the acute hypotensive episode; and
reporting the predicted acute hypotensive episode.
11. The non-transitory computer readable medium according to claim 10 , wherein the method further comprises prohibiting predicting the acute hypotensive episode when the mean arterial pressure is above a predetermined level for a predetermined period.
12. The non-transitory computer readable medium according to claim 10 , wherein the method further comprises filtering data associated with determining heart rate and mean arterial pressure.