Method and apparatus for predicting a need for a blood transfusion

• US 10,258,292 B2
• Filed: 08/12/2014
• Issued: 04/16/2019
• Est. Priority Date: 08/12/2013
• Status: Active Grant

• Current Assignee: University of Maryland Baltimore County (University of Maryland)
• Original Assignee: University of Maryland Baltimore County (University of Maryland)
• Inventors: Hu, Fu-Ming, Mackenzie, Colin, Yang, Shiming, Chen, Hegang
• Primary Examiner(s): Winakur, Eric
• Assistant Examiner(s): Liu, Chu Chuan (JJ)
• Application Number:
  US14/911,421
• Publication Number:
  US 20160183885A1
• Time in Patent Office: 1,708 Days
• CPC Class Codes: A61B 5/02042 Determining blood loss or bleeding, e.g. during a surgical procedure
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/02416 using photoplethysmograph signals, e.g. generated by infrared radiation A61B5/14552 takes precedence
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 5/742 using visual displays A61B5/339 takes precedence
• CPC Class Codes: G16H 20/40 relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A method is provided for predicting that a patient will require a blood transfusion during a treatment. The method includes obtaining, on a processor, first data that indicates values for one or more parameters of a characteristic of a continuous photoplethysmographic (PPG) waveform collected during the treatment. The method further includes applying, on the processor, coefficients to the values for the one or more parameters. The method further includes determining, on the processor, second data that indicates a prediction that the patient will require the blood transfusion during the treatment based on applying the coefficients to the values for the one or more parameters. An apparatus is also provided for predicting that the patient will require the blood transfusion during the treatment.

Claims

1. A method comprising:

obtaining, on a processor, first data that indicates values for one or more parameters of a characteristic of a continuous photoplethysmographic (PPG) waveform collected during a time interval of a treatment of a patient;

applying, on the processor, coefficients to the values for the one or more parameters;

determining, on the processor, second data that indicates a prediction that the patient will require a blood transfusion during a time range within a time period from 15 minutes to 24 hours after the time interval of the treatment based on applying the coefficients to the values for the one or more parameters; and

ordering one or more blood units for a medical facility based on the prediction so that the patient receives a transfusion of the one or more blood units during the time range at the medical facility;

wherein the prediction is based on the time range and wherein the coefficients to determine the prediction based on the time range are in a range from −

100 to 35.

2. The method of claim 1, further comprising presenting on a display device output data based on the second data, wherein the output data indicates the one or more units of blood to be ordered.

3. The method of claim 1, wherein the first data is collected over a fixed time interval, wherein the characteristic of the PPG waveform includes one or more of a heart rate and an oxygen saturation;

and wherein the parameters include one or more of a percentage of the fixed time interval that the heart rate is below a threshold heart rate, a percentage of the fixed time interval that the oxygen saturation is below a threshold saturation rate, a first percentile of the oxygen saturation and a second percentile of the oxygen saturation rate over the fixed time interval, wherein the second percentile is greater than the first percentile.

4. The method of claim 3, wherein the threshold heart rate is 60 beats per minute, the threshold saturation rate is 95%, the first percentile of the oxygen saturation is 25 percentile and the second percentile of the oxygen saturation is 50 percentile.

5. The method of claim 3, wherein the prediction is based on the time range of up to 3 hours after the collection of the first data, wherein the threshold heart rate is 60 beats per minute, the threshold saturation rate is 95%, the first percentile is 25 percentile and the second percentile is 50 percentile;

and wherein the coefficients to determine the prediction based on the time range comprise the coefficient for the percentage of the fixed time interval that the heart rate is below the threshold heart rate in a range from 0.84 to 4.93, the coefficient for the percentage of the fixed time interval that the oxygen saturation is below the threshold saturation rate in a range from 0.05 to 3.45, the coefficient for the first percentile in a range from 0.33 to 2.68 and the coefficient for the second percentile in a range from 0.04 to 2.11.

6. The method of claim 3, wherein the prediction is based on the time range of up to 6 hours after the collection of the first data, wherein the threshold heart rate is 60 beats per minute, the threshold saturation rate is 86%, the first percentile is 25 percentile and the second percentile is 50 percentile;

and wherein the coefficients to determine the prediction based on the time range comprise the coefficient for the percentage of the fixed time interval that the heart rate is below the threshold heart rate in a range from 0.85 to 5.18, the coefficient for the percentage of the fixed time interval that the oxygen saturation is below the threshold saturation rate in a range from 2.45 to 9.45, the coefficient for the first percentile in a range from 0.41 to 2.93 and the coefficient for the second percentile in a range from 0.01 to 2.20.

7. The method of claim 3, wherein the prediction is based on the time range of up to 12 hours after the collection of the first data, wherein the threshold heart rate is 60 beats per minute, the threshold saturation rate is 95%, the first percentile is 25 percentile and the second percentile is 50 percentile;

and wherein the coefficients to determine the prediction based on the time range comprise the coefficient for the percentage of the fixed time interval that the heart rate is below the threshold heart rate in a range from 0.41 to 2.80, the coefficient for the percentage of the fixed time interval that the oxygen saturation is below the threshold saturation rate in a range from 0.04 to 0.26, the coefficient for the first percentile in a range from −

0.11 to 1.72 and the coefficient for the second percentile in a range from −

0.05 to 1.81.

8. The method of claim 3, wherein the prediction is based on the time range of up to 24 hours after the collection of the first data, wherein the threshold heart rate is 60 beats per minute and the threshold saturation rate is 95%;

and wherein the coefficients to determine the prediction based on the time range comprise the coefficient for the percentage of the fixed time interval that the heart rate is below the threshold heart rate in a range from 1.11 to 3.31 and the coefficient for the percentage of the fixed time interval that the oxygen saturation is below the threshold saturation rate in a range from 0.41 to 3.82.

9. The method of claim 1, wherein the first data is collected over a fixed time interval and wherein the parameter is a percentile of an amplitude of the PPG waveform collected over the fixed time interval.

10. The method of claim 9, wherein the parameter is one or more of a 10 percentile, 20 percentile, 25 percentile, 30 percentile, 40 percentile, 50 percentile, 60 percentile, 70 percentile, 75 percentile, 80 percentile, and 90 percentile of the amplitude of the PPG waveform collected over the fixed time interval.

11. The method of claim 1, wherein the PPG waveform includes a heart rate waveform and an oxygen saturation waveform, and wherein the parameter includes one or more of an area of the heart rate waveform below a low threshold heart rate or above a high threshold heart rate and an area of the oxygen saturation waveform below a threshold oxygen saturation rate.

12. The method of claim 11, wherein the low threshold heart rate is one of 60 beats per minute or 72 beats per minute, the high threshold heart rate is one of 100 beats per minute, 110 beats per minute and 120 beats per minute, and the threshold oxygen saturation rate is one or more of 98%, 95%, 92%, 90% and 86%.

13. The method of claim 1 further comprising obtaining, on the processor, third data that indicates values for one or more secondary parameters of a characteristic of the patient;

wherein applying the coefficients to the one or more parameters further includes applying coefficients to the values for the one or more secondary parameters;

and wherein the secondary parameters include one or more of an age and a gender of the patient.

14. The method of claim 1, further comprising determining, on the processor, values for the coefficients of the one or more parameters including:

obtaining, on the processor, preliminary data for the one or more parameters of the characteristic of the PPG waveform during treatment of a plurality of patients;

assigning, on the processor, a result for each patient based on whether the patient received a blood transfusion during the treatment;

fitting, on the processor, the preliminary data to the results for the plurality of patients; and

determining, on the processor, the coefficients for the one or more parameters, to determine a model for predicting whether a patient will require a blood transfusion based on the first data.

15. The method of claim 1, further comprising:

applying, on the processor, secondary coefficients to the values of the one or more parameters; and

determining, on the processor, fourth data that indicates a prediction that the patient will require a massive blood transfusion of at least five blood units during the treatment based on applying the secondary coefficients to the values for the one or more parameters,

wherein the output data is further based on the fourth data;

wherein the secondary coefficients are in a range from −

505 to 128.

16. A method comprising:

obtaining, on a processor, data that indicates values for one or more parameters of a characteristic of a continuous photoplethysmographic (PPG) waveform during a time interval of a treatment of a plurality of patients;

assigning, on the processor, a result for each patient based on whether the patient received a blood transfusion during the treatment;

fitting, on the processor, the data to the results for the plurality of patients;

determining, on the processor, coefficients for the one or more parameters, to determine a model for predicting whether a patient will require a blood transfusion during a time range within a time period from 15 minutes to 24 hours after the time interval based on an input of values for the one or more parameters; and

storing on a computer-readable medium data that indicates the coefficients for the one or more parameters;

wherein the model is for predicting whether the patient will require the blood transfusion within the time range and wherein the coefficients for the model are in a range from −

100 to 35.

17. The method of claim 16,

wherein a result is assigned to each patient during a plurality of time ranges of the treatment based on whether each patient received a blood transfusion during each of the plurality of time ranges;

wherein the data is fit to each respective result for the plurality of patients during the plurality of time ranges;

wherein the coefficients for the one or more parameters are obtained for each of the plurality of time ranges, to determine the model for predicting whether a patient will require a blood transfusion during each of the plurality of time ranges based on an input of values for the one or more parameters.

18. An apparatus comprising:

a pulse oximeter configured to measure first data that indicates values for one or more parameters of a characteristic of a continuous photoplethysmographic (PPG) waveform collected during a time interval of a treatment of a patient;

a display device; and

at least one processor connected to the pulse oximeter to receive the first data of the one or more parameters;

at least one memory including one or more sequence of instructions;

the at least one memory and the one or more sequence of instructions configured to, with the at least one processor, cause the apparatus to perform at least the following;

apply coefficients to the values for the one or more parameters,

determine second data that indicates a prediction that the patient will require a blood transfusion during a time range within a time period from 15 minutes to 24 hours after the time interval of the treatment based on applying the coefficients to the values for the one or more parameters, and

order one or more blood units for a medical facility based on the prediction so that the patient receives a transfusion of the one or more blood units during the time range at the medical facility;

wherein the prediction is based on the time range and wherein the coefficients to determine the prediction based on the time range are in a range from −

100 to 35.

19. A non-transitory computer-readable medium carrying one or more sequences of instructions, wherein execution of the one or more sequences of instructions by one or more processors causes the one or more processors to perform the steps of:

applying coefficients to values for one or more parameters of a characteristic of a continuous photoplethysmographic (PPG) waveform collected during a time interval of a treatment of a patient;

determining a prediction that the patient will require a blood transfusion during a time range within a time period from 15 minutes to 24 hours after the time interval of the treatment based on applying the coefficients to the values for the one or more parameters; and

ordering one or more blood units for a medical facility based on the prediction so that the patient receives a transfusion of the one or more blood units during the time range at the medical facility;

wherein the prediction is based on the time range and wherein the coefficients to determine the prediction based on the time range are in a range from −

100 to 35.