Glycemic health metric determination and application
Disclosed are methods, apparatuses, etc. for determination and application of a metric for assessing a patient's glycemic health. In one particular implementation, a computed metric may be used to balance short-term and long-term risks associated with a particular therapy.
1. An apparatus for use with a patient, with a blood glucose monitoring device, and with an infusion device, the apparatus comprising:
one or more processors configured to:
compute a profile of a blood glucose concentration of the patient based, at least in part, on a probability density function of a plurality of blood glucose concentration values collected at the blood glucose monitoring device;
determine a metric based, at least in part, on a convolution of a cost or loss function and the profile, wherein the metric is for balancing short-term and long-term risks associated with a therapy, and wherein the metric is representative of a glycemic health of the patient, the cost or loss function being based, at least in part, on a log-square operation and being further based, at least in part, on a parameter selected to balance Spearman rank coefficients with hypoglycemic-based metrics and hyperglycemic-based metrics, wherein the log-square operation is applied to at least one of a computed mean of the plurality of blood glucose concentration values or a measure of a statistical dispersion of the plurality of blood glucose concentration values; and
affect an insulin therapy applied to the patient by the infusion device, based, at least in part, on the metric.
2. The apparatus of claim 1 , wherein the measure of the statistical dispersion comprises a standard deviation.
3. The apparatus of claim 1 , wherein to affect the insulin therapy the one or more processors are configured to adjust a target blood glucose level or a target blood glucose range of the patient upward based, at least in part, on an indication by the metric that the patient is tending towards hypoglycemia, or to adjust the target blood glucose level or the target blood glucose range of the patient downward based, at least in part, on an indication by the metric that the patient is tending towards hyperglycemia.
4. The apparatus of claim 1 , wherein the metric is indicative of a portion of time the plurality of blood glucose concentration values has been within a range of 80-180 mg/dl over a duration.
5. The apparatus of claim 1 , wherein the apparatus is further for use with a closed-loop insulin delivery system, and wherein to affect the insulin therapy the one or more processors are configured to affect the closed-loop insulin delivery system based, at least in part, on the metric.
6. An apparatus for use with a patient, with a blood glucose monitoring device, and with an infusion device, the apparatus comprising:
one or more processors configured to:
compute a profile of a blood glucose concentration of the patient based, at least in part, on a probability density function of a plurality of blood glucose concentration values collected at the blood glucose monitoring device;
determine a metric based, at least in part, on a convolution of a cost or loss function and the profile, wherein the metric is for balancing short-term and long-term risks associated with a therapy and is representative of a glycemic health of the patient, and wherein the loss or cost function substantially has the form:
Loss( G T )=log 10 (σ) 2 +[log 10 (μ)−log 10 ( G T )] 2 , where:
G T is a target blood glucose concentration value;
μ is a computed mean of the plurality of blood glucose concentration values; and
σ is a measure of the statistical dispersion of the plurality of blood glucose concentration values; and
affect an insulin therapy applied to the patient by the infusion device, based, at least in part, on the metric.
7. An apparatus for use with a plurality of patients, for use with an insulin delivery system, and for use with one or more blood glucose monitoring devices, the apparatus comprising:
one or more processors configured to:
for each of the plurality of patients, (i) determine a profile of a blood glucose concentration of the patient based, at least in part, on a probability density function of a plurality of blood glucose concentration values collected at the one or more blood glucose monitoring devices, and (ii) determine a metric based, at least in part, on a convolution of a cost or loss function and the profile, wherein the metric is for balancing short-term and long-term risks associated with a therapy, and wherein the metric is representative of a glycemic health of the patient, the cost or loss function being based, at least in part, on a log-square operation and being further based, at least in part, on a parameter selected to balance Spearman rank coefficients with hypoglycemic-based metrics and hyperglycemic-based metrics, wherein the log-square operation is applied to at least one of a computed mean of the plurality of blood glucose concentration values or a measure of a statistical dispersion of the plurality of blood glucose concentration values,
wherein the profiles of the plurality of patients are obtained under multiple predefined therapies, and
wherein the one or more processors are further configured to:
rank the multiple predefined therapies based, at least in part, on the metrics; and control the insulin delivery system for balancing hypoglycemic and hyperglycemic tendencies in the plurality of patients, based at least in part on the ranking of the multiple predefined therapies.
8. The apparatus of claim 7 , wherein the apparatus is further for use with computing platforms co-located with the plurality of patients and for use with a communication network, wherein the apparatus further comprises communication interface components to receive messages from the communication network, and wherein to determine the metric the one or more processors are configured to determine the metric based, at least in part, on messages received through the communication interface components from the computing platforms.
9. The apparatus of claim 8 , wherein the one or more blood glucose monitoring devices comprise a plurality of blood glucose monitoring devices, and wherein said messages comprise measurements of blood glucose concentration collected at the plurality of blood glucose monitoring devices.
10. The apparatus of claim 7 , wherein the multiple predefined therapies are defined, at least in part, by closed-loop system design features.
11. The apparatus of claim 7 , wherein the measure of the statistical dispersion comprises a standard deviation.
12. The apparatus of claim 7 , wherein the metric is indicative of a portion of time the plurality of blood glucose concentration values has been within a predetermined range over a duration.
13. An apparatus for use with a plurality of patients, for use with an insulin delivery system, and for use with one or more blood glucose monitoring devices, the apparatus comprising:
one or more processors configured to:
for each of the plurality of patients, (i) determine a profile of a blood glucose concentration of the patient based, at least in part, on a probability density function of a plurality of blood glucose concentration values collected at the one or more blood glucose monitoring devices, and (ii) determine a metric based, at least in part, on a convolution of a cost or loss function and the profile, wherein the metric is for balancing short-term and long-term risks associated with a therapy and is representative of a glycemic health of the patient, wherein the profiles of the plurality of patients are obtained under multiple predefined therapies, and wherein the loss or cost function substantially has the form:
Loss( G T )=log 10 (σ) 2 +[log 10 (μ)−log 10 ( G T )] 2 , where:
G T is a target blood glucose concentration value;
μ is a computed mean of the plurality of blood glucose concentration values; and
σ is a measure of the statistical dispersion of the plurality of blood glucose concentration values; and
wherein the one or more processors are further configured to:
rank the multiple predefined therapies based, at least in part, on the metrics; and control the insulin delivery system for balancing hypoglycemic and hyperglycemic tendencies in the plurality of patients, based at least in part on the ranking of the multiple predefined therapies.
14. An article for use with a patient, with a blood glucose monitoring device, and with an infusion device, the article comprising:
a non-transitory storage medium comprising machine-readable instructions stored thereon which are executable by a special purpose computing apparatus to:
compute a profile of a blood glucose concentration of the patient based, at least in part, on a probability density function of a plurality of blood glucose concentration values collected at the blood glucose monitoring device;
determine a metric based, at least in part, on a convolution of a cost or loss function and the profile, wherein the metric is for balancing short-term and long-term risks associated with a therapy, and wherein the metric is representative of a glycemic health of the patient, the cost or loss function being based, at least in part, on a log-square operation and being further based, at least in part, on a parameter selected to balance Spearman rank coefficients with hypoglycemic-based metrics and hyperglycemic-based metrics, wherein the log-square operation is applied to at least one of a computed mean of the plurality of blood glucose concentration values or a measure of a statistical dispersion of the plurality of blood glucose concentration values; and
affect an insulin therapy applied to the patient by the infusion device based, at least in part, on the metric.
15. An article for use with a plurality of patients, for use with an insulin delivery system, and for use with one or more blood glucose monitoring devices, the article comprising:
a non-transitory storage medium comprising machine-readable instructions stored thereon which are executable by a special purpose computing apparatus to:
for each of the plurality of patients, (i) determine a profile of a blood glucose concentration of the patient based, at least in part, on a probability density function of a plurality of blood glucose concentration values collected at the one or more blood glucose monitoring devices, and (ii) determine a metric based, at least in part, on a convolution of a cost or loss function and the profile, wherein the metric is for balancing short-term and long-term risks associated with a therapy, and wherein the metric is representative of a glycemic health of the patient, the cost or loss function being based, at least in part, on a log-square operation and being further based, at least in part, on a parameter selected to balance Spearman rank coefficients with hypoglycemic-based metrics and hyperglycemic-based metrics, wherein the log-square operation is applied to at least one of a computed mean of the plurality of blood glucose concentration values or a measure of a statistical dispersion of the plurality of blood glucose concentration values,
wherein the profiles of the plurality of patients are obtained under multiple predefined therapies, and
wherein the machine-readable instructions are further executable by the special purpose computing apparatus to:
rank the multiple predefined therapies based, at least in part, on the metrics; and control the insulin delivery system for balancing hypoglycemic and hyperglycemic tendencies in the plurality of patients, based at least in part on the ranking of the multiple predefined therapies.