Default carbohydrate consumption counts based on population carbohydrate consumption models
Techniques disclosed herein relates generally to default carbohydrate consumption counts based on population carbohydrate consumption models. In one example, the techniques involve receiving, from a device, at least one characteristic of a person; accessing a population carbohydrate consumption model which relates the at least one characteristic with carbohydrate intake for a population; determining, based on applying the population carbohydrate consumption model to the at least one characteristic, at least one of: a preliminary carbohydrate count for the person or a default carbohydrate consumption count for the person; and causing delivery of insulin to the person based on communicating, to the device, at least one of: the preliminary carbohydrate count for the person or the default carbohydrate consumption count for the person.
1 . A system comprising:
one or more processors; and
one or more memory devices storing instructions which, when executed by the one or more processors, cause performance of:
receiving, from a device, at least one characteristic of a person that is different from glucose measurement data of the person, the at least one characteristic being identified in a personal profile being stored on the device;
selecting a population carbohydrate consumption model from among a plurality of population carbohydrate consumption models based at least in part on the at least one characteristic identified in the personal profile, wherein the population carbohydrate consumption model relates the at least one characteristic with an amount of carbohydrate intake for a population;
determining, based on applying the population carbohydrate consumption model to the at least one characteristic, at least one of: a preliminary carbohydrate count for the person or a default carbohydrate consumption count for the person; and
causing delivery of insulin to the person based on communicating, to the device, at least one of: the preliminary carbohydrate count for the person or the default carbohydrate consumption count for the person.
2 . The system of claim 1 , wherein the default carbohydrate consumption count is determined based on the preliminary carbohydrate count and a scaling factor.
3 . The system of claim 2 , wherein the scaling factor is determined based on glycemic response simulations for simulated persons.
4 . The system of claim 1 , wherein:
the at least one characteristic of the person includes at least one of: age of the person or body weight of the person, and
the population carbohydrate consumption model relates at least one of age or body weight with carbohydrate intake for the population.
5 . The system of claim 1 , wherein the population carbohydrate consumption model is generated based on machine learning or regression analysis.
6 . The system of claim 5 , wherein the machine learning involves a deep learning neural network.
7 . A processor-implemented method comprising:
receiving, from a device, at least one characteristic of a person that is different from glucose measurement data of the person, the at least one characteristic being identified in a personal profile being stored on the device;
selecting a population carbohydrate consumption model from among a plurality of population carbohydrate consumption models based at least in part on the at least one characteristic identified in the personal profile, wherein the population carbohydrate consumption model relates the at least one characteristic with an amount of carbohydrate intake for a population;
determining, based on applying the population carbohydrate consumption model to the at least one characteristic, at least one of: a preliminary carbohydrate count for the person or a default carbohydrate consumption count for the person; and
causing delivery of insulin to the person based on communicating, to the device, at least one of: the preliminary carbohydrate count for the person or the default carbohydrate consumption count for the person.
8 . The processor-implemented method of claim 7 , wherein the default carbohydrate consumption count is determined based on the preliminary carbohydrate count and a scaling factor.
9 . The processor-implemented method of claim 8 , wherein the scaling factor is determined based on glycemic response simulations for simulated persons.
10 . The processor-implemented method of claim 7 , wherein:
the at least one characteristic of the person includes at least one of: age of the person or body weight of the person, and
the population carbohydrate consumption model relates at least one of age or body weight with carbohydrate intake for the population.
11 . The processor-implemented method of claim 7 , wherein the population carbohydrate consumption model is generated based on machine learning or regression analysis.
12 . The processor-implemented method of claim 11 , wherein the machine learning involves a deep learning neural network.
13 . A non-transitory machine-readable medium storing instructions which, when executed by one or more processors, cause performance of:
receiving, from a device, at least one characteristic of a person that is different from glucose measurement data of the person, the at least one characteristic being identified in a personal profile being stored on the device;
selecting a population carbohydrate consumption model from among a plurality of population carbohydrate consumption models based at least in part on the at least one characteristic identified in the personal profile, wherein the population carbohydrate consumption model relates the at least one characteristic with an amount of carbohydrate intake for a population;
determining, based on applying the population carbohydrate consumption model to the at least one characteristic, at least one of: a preliminary carbohydrate count for the person or a default carbohydrate consumption count for the person; and
causing delivery of insulin to the person based on communicating, to the device, at least one of: the preliminary carbohydrate count for the person or the default carbohydrate consumption count for the person.
14 . The non-transitory machine-readable medium of claim 13 , wherein the default carbohydrate consumption count is determined based on the preliminary carbohydrate count and a scaling factor.
15 . The non-transitory machine-readable medium of claim 14 , wherein the scaling factor is determined based on glycemic response simulations for simulated persons.
16 . The non-transitory machine-readable medium of claim 13 , wherein:
the at least one characteristic of the person includes at least one of: age of the person or body weight of the person, and
the population carbohydrate consumption model relates at least one of age or body weight with carbohydrate intake for the population.
17 . The non-transitory machine-readable medium of claim 13 , wherein the population carbohydrate consumption model is generated based on machine learning or regression analysis.
18 . The method of claim 7 , wherein the at least one characteristic of the person includes at least one of a demographic characteristic, a personal characteristic, a location characteristic, or an activity characteristic of the person.
19 . The method of claim 8 , wherein the scaling factor is a personalized scaling factor for the person.
20 . The method of claim 8 , wherein the default carbohydrate consumption count for the person determined based on the preliminary carbohydrate count and the scaling factor is further bounded by a lower limit and an upper limit.