Statistical, noninvasive measurement of a patient's physiological state
Tools and techniques for the rapid, continuous, invasive and/or noninvasive measurement, estimation, and/or prediction of a patient's physiological state. In an aspect, some tools and techniques can estimate predict the onset of conditions intracranial pressure, an amount of blood volume loss, cardiovascular collapse, and/or dehydration. Some tools can recommend (and, in some cases, administer) a therapeutic treatment for the patient's condition. In another aspect, some techniques employ high speed software technology that enables active, long term learning from extremely large, continually changing datasets. In some cases, this technology utilizes feature extraction, state-of-the-art machine learning and/or statistical methods to autonomously build and apply relevant models in real-time.
1. A computational method of estimating a patient's physiological state, the method comprising:
generating a model of the physiological state, wherein generating the model comprises:
receiving data pertaining to a plurality of physiological parameters of a test subject to obtain a plurality of physiological data sets;
directly measuring a physiological state of the test subject with a reference sensor to obtain a plurality of physiological state measurements; and
correlating the received data with the physiological state of the test subject, wherein correlating the received data with the measured physiological state comprises:
identifying a most-predictive set of signals S k out of a set of signals s 1 , s 2 , . . . , s D for each of one or more outcomes o k , wherein the most-predictive set of signals S k corresponds to a first data set representing a first physiological parameter, and wherein each of the one or more outcomes o k represents one of the plurality of intracranial pressure measurements;
autonomously learning a set of probabilistic predictive models ô k =M k (S k ), where ô k is a prediction of outcome o k derived from a model M k that uses as input values obtained from the set of signals S k ; and
repeating the operation of autonomously learning incrementally from data that contains examples of values of signals s 1 , s 2 , . . . , s D and corresponding outcomes o 1 , o 2 , . . . , o K ;
receiving, at a computer system, a set of input data from one or more physiological sensors, the input data pertaining to one or more physiological parameters of a patient;
analyzing, with the computer system, the input data against the model to generate diagnostic data concerning the patient's physiological state; and
displaying, with a display device, at least a portion of the diagnostic data concerning the patient's physiological state.
2. The method of claim 1 , wherein the patient's physiological state comprises intracranial pressure, and wherein the diagnostic data comprises an estimate of a current intracranial pressure of the patient.
3. The method of claim 1 , wherein the model is a hemodynamic compensation model, and wherein the patient's physiological state is a volume of acute blood loss.
4. The method of claim 3 , wherein the diagnostic data concerning the patient's physiological state comprises a prediction of when the patient will experience cardiovascular collapse.
5. The method of claim 1 , wherein the physiological state is a state of dehydration.
6. The method of claim 1 , wherein the method is noninvasive to the patient, and wherein the one or more physiological sensors are noninvasive sensors.
7. The method of claim 1 , further comprising:
controlling, with the computer system, a therapeutic treatment of the patient's physiological state, based at least in part on the generated diagnostic data.
8. The method of claim 7 , wherein controlling a therapeutic treatment comprises selecting one or more recommended treatment options, out of a plurality of treatment options.
9. The method of claim 8 , wherein controlling a therapeutic treatment further comprises administering, with the computer system, the one or more recommended treatment options.
10. The method of claim 1 , further comprising:
repeating the operation of receiving a set of input data over a plurality of iterations, to produce a plurality of sets of input data, wherein analyzing the input data comprises analyzing the plurality of sets of input data to generate a plurality of sets of diagnostic data;
wherein displaying at least a portion of the diagnostic data comprises updating a display on the display device to display at least some of the plurality of sets of diagnostic data over the period of time.
11. The method of claim 10 , wherein the plurality of sets of input data comprise periodic samples of a set of continuous, physiological waveform data.
12. The method of claim 11 , wherein the set of continuous, physiological waveform data comprises blood pressure waveform data.
13. The method of claim 1 , wherein the one or more physiological sensors comprise an electrocardiograph sensor.
14. The method of claim 1 , wherein the one or more physiological sensors comprise an impedance cardiograph sensor.
15. The method of claim 1 , wherein the one or more physiological sensors comprise a pulse oximeter.
16. The method of claim 1 , wherein the one or more physiological sensors comprise a plethysmograph.
17. The method of claim 1 , wherein the one or more physiological sensors comprise a photoplethysmograph (PPG).
18. The method of claim 1 , wherein the one or more physiological sensors comprise a near infrared sensor.
19. The method of claim 1 , wherein the one or more physiological sensors comprise a transcranial Doppler sensor.
20. The method of claim 1 , wherein autonomously learning the set of probabilistic predictive models comprises using a linear model framework to identify predictive variables for each increment of data.
21. The method of claim 20 , wherein the linear model framework is constructed with the form
o
^
k
=
f
k
(
a
0
+
∑
i
=
1
d
a
i
s
i
)
,
where f k is any mapping from one input to one output and a 0 , a 1 , . . . , a d are linear model coefficients.
22. The method of claim 1 , further comprising:
updating the model using the set of input data.
23. An apparatus, comprising:
a non-transitory computer readable medium having encoded thereon a set of instructions executable by one or more computers to:
generate a model of the physiological state, wherein generating the model comprises:
receiving data pertaining to a plurality of physiological parameters of a test subject to obtain a plurality of physiological data sets;
directly measuring a physiological state of the test subject with a reference sensor to obtain a plurality of physiological state measurements; and
correlating the received data with the physiological state of the test subject, wherein correlating the received data with the measured physiological state comprises:
identifying a most-predictive set of signals S k out of a set of signals s 1 , s 2 , . . . , s D for each of one or more outcomes o k , wherein the most-predictive set of signals S k corresponds to a first data set representing a first physiological parameter, and wherein each of the one or more outcomes o k represents one of the plurality of intracranial pressure measurements;
autonomously learning a set of probabilistic predictive models ô k =M k (S k ), where o k is a prediction of outcome o k derived from a model M k that uses as input values obtained from the set of signals S k ; and
repeating the operation of autonomously learning incrementally from data that contains examples of values of signals s 1 , s 2 , . . . , s D and corresponding outcomes o 1 , o 2 , . . . , o K ;
receive a set of input data from one or more physiological sensors, the input data pertaining to one or more physiological parameters of a patient;
analyze the input data against the model to generate diagnostic data concerning the patient's physiological state; and
display, with a display device, at least a portion of the diagnostic data concerning the patient's physiological state.
24. A system, comprising:
a computer system, comprising:
one or more processors; and
a non-transitory computer readable medium in communication with the one or more processors, the computer readable medium having encoded thereon a set of instructions executable by the computer system to:
generate a model of the physiological state, wherein generating the model comprises:
receiving data pertaining to a plurality of physiological parameters of a test subject to obtain a plurality of physiological data sets;
directly measuring a physiological state of the test subject with a reference sensor to obtain a plurality of physiological state measurements; and
correlating the received data with the physiological state of the test subject, wherein correlating the received data with the measured physiological state comprises:
identifying a most-predictive set of signals S k out of a set of signals s 1 , s 2 , . . . , s D for each of one or more outcomes Ok, wherein the most-predictive set of signals S k corresponds to a first data set representing a first physiological parameter, and wherein each of the one or more outcomes o k represents one of the plurality of intracranial pressure measurements;
autonomously learning a set of probabilistic predictive models ô k =M k (S k ), where ô k is a prediction of outcome o k derived from a model M k that uses as input values obtained from the set of signals S k ; and
repeating the operation of autonomously learning incrementally from data that contains examples of values of signals s 1 , s 2 , . . . , s D and corresponding outcomes o 1 , o 2 , . . . , o K ;
receive a set of input data from one or more physiological sensors, the input data pertaining to one or more physiological parameters of a patient;
analyze the input data against the model to generate diagnostic data concerning the patient's physiological state; and
display, with a display device, at least a portion of the diagnostic data concerning the patient's physiological state.
25. The system of claim 23 , further comprising the one or more physiological sensors.