Blood purification apparatus, blood purification method, program, and model generating method

Abstract

A blood purification device (10) comprises: a measurement unit (44) that uses a cuff (54) for pressurizing a blood vessel of a patient undergoing blood purification treatment to measure the patient's blood pressure values and the patient's pulse wave amplitude values corresponding to cuff pressures; and a prediction unit (50) that receives, as input, time-series data including the blood pressure values and the pulse wave amplitude values measured in the blood purification treatment, and outputs the patient's blood pressure prediction value.

Claims

1. A blood purification device comprising: a measuring unit that uses a cuff to pressurize the blood vessels of a patient undergoing blood purification therapy to measure the patient's blood pressure and pulse wave amplitude value relative to the cuff pressure; and a predictive unit that takes time-series data including the blood pressure and pulse wave amplitude values ​​measured during the blood purification therapy as input and outputs a predicted value for the patient's blood pressure.

2. The blood purification apparatus according to claim 1, wherein the prediction unit takes time-series data including the blood pressure value and pulse wave amplitude value measured at two or more points in time during the blood purification treatment as input and outputs a predicted blood pressure value of the patient at a point later than the two or more points in time during the blood purification treatment.

3. The blood purification device according to claim 1 or 2, wherein the prediction unit outputs a predicted time at which a decrease in blood pressure occurs in the patient during the blood purification treatment.

4. The blood purification device according to any one of claims 1 to 3, wherein the prediction unit further uses the day of the week on which the blood purification treatment is performed, or the number of days elapsed since the previous blood purification treatment for the patient, as input.

5. The blood purification device according to any one of claims 1 to 4, further comprising a notification unit that notifies the user based on the output result of the prediction unit.

6. The blood purification apparatus according to any one of claims 1 to 5, further comprising a display unit for displaying the output results of the prediction unit.

7. The blood purification apparatus according to any one of claims 1 to 6, further comprising a dialysis control unit that changes the ultrafiltration rate of the blood purification treatment, stops ultrafiltration, or performs emergency fluid replacement based on the output result of the prediction unit.

8. The blood purification apparatus according to any one of claims 1 to 7, wherein the prediction unit uses a prediction model generated by machine learning of a plurality of time series data including blood pressure values ​​and pulse wave amplitude values ​​measured in past blood purification treatments for the patient.

9. A blood purification method comprising the steps of: measuring the patient's blood pressure and pulse wave amplitude value relative to the cuff pressure using a cuff that pressurizes the blood vessels of the patient undergoing blood purification therapy; and outputting a predicted blood pressure value for the patient during the blood purification therapy, using time-series data including the blood pressure and pulse wave amplitude values ​​measured during the blood purification therapy as input.

10. A program that enables a computer to perform the following functions: acquiring the patient's blood pressure value and pulse wave amplitude value in relation to the cuff pressure, measured using a cuff that pressurizes the blood vessels of the patient undergoing blood purification therapy; and outputting a predicted blood pressure value for the patient during the blood purification therapy, taking time-series data including the blood pressure value and pulse wave amplitude value measured during the blood purification therapy as input.

11. A model generation method comprising the steps of: acquiring time-series data including a patient's blood pressure and pulse wave amplitude measured at multiple points in time during blood purification therapy; and generating a predictive model using machine learning that outputs a predicted value of the patient's blood pressure during the blood purification therapy using the time-series data.

Images