Apparatus and method for generating a condition indication

• US 9,336,355 B2
• Filed: 02/10/2012
• Issued: 05/10/2016
• Est. Priority Date: 08/10/2009
• Status: Expired due to Fees

• Current Assignee: Diabetes Tools Sweden AB
• Original Assignee: Diabetes Tools Sweden AB
• Inventors: Ljuhs, Michael Kjell, Riback, Jacob Lars Fredrik, Liljeryd, Lars Gustaf
• Primary Examiner(s): Lin, Jerry
• Application Number:
  US13/370,699
• Publication Number:
  US 20120209099A1
• Time in Patent Office: 1,551 Days
• Field of Search: 702/19
• US Class Current: 1/1
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/4839 combined with drug delivery
• CPC Class Codes: A61B 5/7253 characterised by using transforms
• 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 5/746 Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms alarms per se G08B
• CPC Class Codes: A61M 2230/005 Parameter used as control input for the apparatus
• CPC Class Codes: A61M 2230/201 Glucose concentration
• CPC Class Codes: A61M 5/1723 using feedback of body parameters, e.g. blood-sugar, pressure measurement of body parameters A61B5/00
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 20/10 relating to drugs or medications, e.g. for ensuring correct administration to patients
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

An apparatus for generating a condition indication using a time sequence of data values, each data value representing a physiological measure of a condition of a subject at a time, includes: a transformer for transforming the time sequence of data values into a transformed sequence of data values using a transform rule, wherein the transform rule is such that a certain characteristic in a time course of the physiological measure is represented by the transformed sequence of data values is more linear than the time course before the transform; a rate of change calculator for calculating an estimated rate of change for the transformed sequence of data values; and a processor for processing the estimated rate of change to output the condition indication.

Claims

1. Apparatus for generating a condition indication using a time sequence of data values, each data value representing a physiological measure of a condition of a subject at a time, wherein the physiological measure is a blood glucose value comprising:

a transformer configured to transform the time sequence of data values into a transformed sequence of data values using a transform rule, wherein the transform rule is such that a certain characteristic in a time course of the physiological measure as represented by the transformed sequence of data values is more linear than the time course before the transform, said transformer comprising;

an estimated probability function calculator configured to calculate an estimated cumulative distribution function associated with a set of data values; and

a transform calculator configured to calculate the transform rule using a predetermined linearly increasing target cumulative distribution function indicating the same probability for a certain data value within the certain characteristic;

a rate of change calculator configured to calculate an estimated rate of change for the transformed sequence of data values; and

a processor configured to process the estimated rate of change to output the condition indication,

wherein at least one of the transformer, rate of change calculator and processor include a hardware implementation.

2. Apparatus in accordance with claim 1, in which the processor is configured to generate a graphic, audible or tactile display of the estimated rate of change or to generate an electrical, magnetic or electro-magnetic signal representing the estimated rate of change.

3. Apparatus in accordance with claim 1, in which the transformed sequence of data values has a transformed value for a current time instant, and wherein the processor is configured to predict a data value for a later time instance from the estimated rate of change, in which the processor is configured to evaluate, whether the predicted data value exceeds a high alarm threshold or is below a low alarm threshold, and in which the processor is configured to generate a visual, audible, tactile, mechanical, electric or magnetic alarm indication.

4. Apparatus in accordance with claim 1, in which the transformer comprises a memory configured to store the transform rule as parameters for a parameterized function, or as a look-up table in which non-transform data values have associated transformed data values.

5. Apparatus in accordance with claim 1, in which the time sequence of data values comprises glycemic data values indicating a glycemic condition for a time instance of the subject.

6. Apparatus in accordance with claim 1,

in which the certain characteristic is represented by a characteristic data record, the characteristic data record comprising value data derived by average rates of change for different value ranges in a measurement range, where the value data have associated time instants, and

in which the rate of change calculator is configured to calculate a difference between two transformed values of the sequence of transformed values and to divide the difference by a time difference calculated between the time instances associated with the two transformed values.

7. Apparatus in accordance with claim 3, in which the processor is configured to calculate the predicted data value using a linear function defined by a data value of the time sequence of data values and the calculated rate of change.

8. Apparatus in accordance with claim 7,

in which the processor is configured to calculate a data value defined by the linear function at a later time being a predetermined time period later than the data value defining the linear function, or

in which the processor is configured to calculate a later time value of an intersection of the linear function and the high or low alarm value and to output an indication, when the condition of the subject approaches an alarm state based on the later time value.

9. Apparatus in accordance with claim 1, in which the estimated probability function calculator is configured to update the estimated cumulative distribution function by averaging, for a plurality of data level regions, the rate of change values represented by the time sequence of data values to obtain an updated average rate of change for each data value region, and in which the transform calculator is configured to calculate the non-linear transform rule using the data with the updated average rates of change for the data value regions, wherein the same target cumulative distribution function is used.

10. Apparatus in accordance with claim 9, in which the estimated probability function calculator is configured to calculate a plurality of k normal distributions for a plurality of k regions, where each region represents a value range, and wherein adjacent regions overlap each other, so that, for each region, a Gaussian normal probability distribution is estimated using values in the corresponding regions, and wherein the estimated probability function calculator is configured to calculate weights for each region, so that an integral over the sum of the k weighted normal probability distributions results in unity.

11. Apparatus in accordance with claim 1, further comprising a transform rule calculator being configured to calculate a time course of the certain characteristic, as an initial calculation or as a re-calculation in response to a certain re-calculation event, by selecting from measurement data of an individual or a plurality of individuals portions comprising the characteristic, by dividing a whole value range into a plurality of different regions, by calculating, for each region, an average rate of change, and by building a characteristic dynamic behaviour curve from the average rate of change values for the regions, wherein the time course of the certain characteristic is based on the characteristic dynamic behaviour curve.

12. Apparatus in accordance with claim 1, in which the transformer comprises

an estimated characteristic function calculator configured to calculate a time course of the certain characteristic and a target function calculator configured to calculate a target function as a linear function determined by two points of the time course of the certain characteristic;

and

a transform calculator configured to calculate the transform rule using the target function and the estimated characteristic function so that transformed data is closer to the target function than to the estimated characteristic function.

13. Apparatus in accordance with claim 1, wherein the certain characteristic is a blood glucose rise or a blood glucose fall.

14. Apparatus in accordance with claim 1, wherein the apparatus further comprises a device for dosing a medication, and wherein the processor is further configured to control the device for dosing a medication based at least in part on one of the transformed data values.

15. Apparatus in accordance with claim 14, wherein the device for dosing a medication includes an insulin pump.

16. Method for generating a condition indication using a time sequence of data values, each data value representing a physiological measure of a condition of a subject at a time, wherein the physiological measure is a blood glucose value, comprising:

transforming the time sequence of data values into a transformed sequence of data values using a transform rule, wherein the transform rule is such that a certain characteristic in a time course of the physiological measure as represented by the transformed sequence of data values is more linear than the time course before the transform, said transforming comprising;

calculating an estimated cumulative distribution function associated with a set of data values; and

calculating the transform rule using a predetermined linearly increasing target cumulative distribution function indicating the same probability for a certain data value within the certain characteristic;

calculating an estimated rate of change for the transformed sequence of data values; and

processing the estimated rate of change to output the condition indication.

17. A non-transitory computer-readable storage medium having stored thereon a computer program having a program code for performing, when running on a computer or processor, the method for generating a condition indication using a time sequence of data values, each data value representing a physiological measure of a condition of a subject at a time, wherein the physiological measure is a blood glucose value, comprising:

transforming the time sequence of data values into a transformed sequence of data values using a transform rule, wherein the transform rule is such that a certain characteristic in a time course of the physiological measure as represented by the transformed sequence of data values is more linear than the time course before the transform, said transforming comprising;

calculating an estimated cumulative distribution function associated with a set of data values; and

calculating the transform rule using a predetermined linearly increasing target cumulative distribution function indicating the same probability for a certain data value within the certain characteristic;

calculating an estimated rate of change for the transformed sequence of data values; and

processing the estimated rate of change to output the condition indication.

18. Method for controlling a device for dosing a medication using a sequence of data values, each data value representing a physiological measure of a condition of a subject at a time, wherein the physiological measure is a blood glucose value, comprising:

transforming the time sequence of data values into a transformed sequence of data values using a transform rule, wherein the transform rule is such that a certain characteristic in a time course of the physiological measure as represented by the transformed sequence of data values is more linear than the time course before the transform, said transforming comprising;

calculating an estimated cumulative distribution function associated with a set of data values; and

calculating the transform rule using a predetermined linearly increasing target cumulative distribution function indicating the same probability for a certain data value within the certain characteristic; and

controlling the device for dosing a medication based at least in part on one of the transformed data values.