Rapid detection of bleeding before, during, and after fluid resuscitation
Novel tools and techniques are provided for assessing, predicting and/or estimating a probability that a patent is bleeding, in some cases, noninvasively. In various embodiments, tools and techniques are provided for implementing rapid detection of bleeding before, during, and after fluid resuscitation, in some instances, in real-time.
1. A method, comprising:
monitoring, with one or more sensors, physiological data of a patient over a time period, wherein the physiological data is cardiovascular data of the patient;
analyzing, with a computer system, the physiological data, wherein analyzing the physiological data further comprises:
comparing a sample of the physiological data against a model, the model comprising reference data representing a plurality of reference waveforms, each reference waveform of the plurality of reference waveforms corresponding to a respective compensatory reserve index (“CRI”) value based on at least on a ratio of a time-dependent intravascular volume loss and an intravascular volume loss at a point of hemodynamic decompensation,
wherein the compared, sampled physiological data comprises data representing one or more patient waveforms obtained over the time period,
wherein comparing the sampled physiological data to the model further comprises comparing the sampled physiological data representing each of the one or more patient waveforms with the reference data representing each of the plurality of reference waveforms of the model;
determining a similarity between the compared, sampled physiological data of each of the one or more patient waveforms and the reference data representing each of the plurality of reference waveforms of the model;
estimating a plurality of CRI values of the patient, based at least in part on the determined similarities;
estimating a probability that the patient is bleeding based on, at least in part, a rate of change between at least two of the plurality of values of CRI that are estimated by comparison of the physiological data obtained from the patient at two or more respective times during the time period to the reference data representing each of the plurality of reference waveforms of the model; and
outputting one or more control signals to control the operation of a rapid transfuser to transfuse the patient with blood or blood products based on the assessment that the patient is probably bleeding.
2. The method of claim 1 further comprising displaying, on a display device, an indicator of the estimate of the probability that the patient is bleeding, wherein one or more of (i) monitoring the physiological data, (ii) analyzing the physiological data, (iii) estimating the probability that the patient is bleeding, or (iv) displaying the indicator in real-time.
3. The method of claim 1 , wherein the physiological data obtained from the patient at the two or more respective times includes physiological data obtained at least one of before, during, or after a fluid resuscitation procedure.
4. The method of claim 1 , wherein estimating the probability that the patient is bleeding further comprises estimating the probability that the patient is bleeding based at least in part on an average value of CRI over the period of time.
5. The method of claim 1 , wherein estimating the probability that the patient is bleeding further comprises estimating the probability that the patient is bleeding based at least in part on a standard deviation of at least some of the plurality of values of CRI.
6. The method of claim 1 , wherein estimating the probability that the patient is bleeding further comprises estimating the probability that the patient is bleeding based at least in part on a skewness of at least some of the plurality of values of CRI.
7. The method of claim 2 , wherein the indicator is a value between 0 and 1.
8. The method of claim 7 , wherein a value of 0 indicates that the patient is not bleeding, and wherein a value of 1 indicates that the patient is bleeding.
9. The method of claim 1 , wherein the one or more reference waveforms correspond to a state of hemodynamic decompensation or near hemodynamic decompensation, or a series of states progressing towards hemodynamic decompensation.
10. The method of claim 1 , wherein determining the similarity further comprises:
producing, for each respective patient waveform of the one or more patient waveforms obtained over the time period, a respective similarity coefficient representing a similarity between a respective patient waveform and each of the plurality of reference waveforms of the model;
normalizing the produced, respective similarity coefficients;
summing each respective CRI value, corresponding to each of the one or more reference waveforms of the model, respectively, weighted by the normalized, produced similarity coefficient; and
determining an estimated CRI value for the patient based on the summed, weighted CRI values.
11. The method of claim 1 , wherein the time period comprises a fixed, historical monitoring time period of the patient.
12. The method of claim 1 , wherein the time period comprises a dynamic, historical monitoring time period of the patient.
13. The method of claim 1 , wherein at least one of the one or more sensors is selected from the group consisting of a blood pressure sensor, an intracranial pressure monitor, a central venous pressure monitoring catheter, an arterial catheter, an electroencephalograph, a cardiac monitor, a transcranial Doppler sensor, a transthoracic impedance plethysmograph, a pulse oximeter, a near infrared spectrometer, a ventilator, an accelerometer, and an electronic stethoscope.
14. The method of claim 1 , wherein the physiological data comprises data representing blood pressure waveforms.
15. The method of claim 1 , wherein the physiological data comprises data representing plethysmograph waveforms.
16. The method of claim 1 , wherein the physiological data comprises data representing photoplethysmograph (PPG) waveforms.
17. The method of claim 1 , wherein the plurality of reference waveforms are generated by inducing a test subject to enter one or more physiological states, and obtaining physiological data from the test subject while the test subject is in a respective physiological state of one or more physiological states.
18. The method of claim 17 , further comprising: inducing a physiological state of reduced circulatory system volume in the test subject.
19. The method of claim 18 , wherein inducing the physiological state comprises subjecting the test subject to lower body negative pressure (“LBNP”).
20. The method of claim 19 , wherein inducing the physiological state comprises subjecting the test subject to dehydration.
21. The method of claim 17 , wherein the one or more physiological states comprises a state of cardiovascular collapse or near-cardiovascular collapse.
22. The method of claim 17 , wherein the one or more physiological states comprises a state of euvolemia.
23. The method of claim 17 , wherein the one or more physiological states comprises a state of hypervolemia.
24. The method of claim 17 , wherein the one or more physiological states comprises a state of dehydration.
25. The method of claim 17 , further comprising correlating the physiological data of the test subject to the respective physiological state, wherein correlating the physiological data of the test subject with the physiological state of the test subject 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 of the physiological data of the test subject, and wherein each of the one or more outcomes o k represents a physiological state measurement of the one or more physiological states respectively;
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 inputs 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 .
26. A 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 one or more processors to cause the computer system to,
monitor, with one or more sensors, physiological data of a patient over a time period, wherein the physiological data is cardiovascular data of the patient;
analyze, with a computer system, the physiological data, wherein the analysis comprises:
a comparison of the physiological data representing one or more patient waveforms obtained over the time period against a model, the model comprising reference data representing a plurality of reference waveforms, each reference waveform of the plurality of reference waveforms corresponding to a respective compensatory reserve index (“CRI”) value based on at least on a ratio of a time-dependent intravascular volume loss and an intravascular volume loss at a point of hemodynamic decompensation,
wherein the comparison of the physiological data to the model further comprises comparison of the physiological data representing each of the one or more patient waveforms with the reference data representing each of the plurality of reference waveforms of the model;
determine a similarity between the compared physiological data of each of the one or more patient waveforms and the reference data representing each of the plurality of reference waveforms of the model;
estimate a plurality of CRI values of the patient, based at least in part on the determined similarities;
estimate a probability that the patient is bleeding based on, at least in part, a rate of change between at least two of the plurality of values of CRI that are estimated by comparison of the physiological data obtained from the patient at two or more respective times during the time period to the reference data representing each of the plurality of reference waveforms of the model; and
output one or more control signals to control the operation of a rapid transfuser to transfuse the patient with blood or blood products based on the assessment that the patient is probably bleeding.
27. An apparatus, comprising:
a non-transitory computer readable medium in communication with one or more processors, the computer readable medium having encoded thereon a set of instructions executable by the one or more processors to cause the computer system to,
monitor, with one or more sensors, physiological data of a patient over a time period, wherein the physiological data is cardiovascular data of the patient;
analyze, with a computer system, the physiological data, wherein the analysis comprises:
a comparison of the physiological data representing one or more patient waveforms obtained over the time period against a model, the model comprising reference data representing a plurality of reference waveforms, each reference waveform of the plurality of reference waveforms corresponding to a respective compensatory reserve index (“CRI”) value based on at least on a ratio of a time-dependent intravascular volume loss and an intravascular volume loss at a point of hemodynamic decompensation,
wherein the comparison of the physiological data to the model further comprises comparison of the physiological data representing each of the one or more patient waveforms with the reference data representing each of the plurality of reference waveforms of the model;
determine a similarity between the compared physiological data of each of the one or more patient waveforms and the reference data representing each of the plurality of reference waveforms of the model;
estimate a plurality of CRI values of the patient, based at least in part on the determined similarities;
estimate a probability that the patient is bleeding based on, at least in part, a rate of change between at least two of the plurality of values of CRI that are estimated by comparison of the physiological data obtained from the patient at two or more respective times during the time period to the reference data representing each of the plurality of reference waveforms of the model; and
output one or more control signals to control the operation of a rapid transfuser to transfuse the patient with blood or blood products based on the assessment that the patient is probably bleeding.