US 12,073,333 B2
Predicting rates of hypoglycemia by a machine learning system
Francisco Javier Jimenez Jimenez, Brookline, MA (US); Hsiaohui Wu, Ringoes, NJ (US); and Fang Liz Zhou, Warren, NJ (US)
Assigned to Sanofi, Paris (FR)
Appl. No. 17/251,091
Filed by Sanofi, Paris (FR)
PCT Filed Jun. 21, 2019, PCT No. PCT/US2019/038454
§ 371(c)(1), (2) Date Dec. 10, 2020,
PCT Pub. No. WO2019/246511, PCT Pub. Date Dec. 26, 2019.
Claims priority of provisional application 62/689,005, filed on Jun. 22, 2018.
Claims priority of application No. 19305609 (EP), filed on May 13, 2019.
Prior Publication US 2021/0216894 A1, Jul. 15, 2021
Int. Cl. G06N 5/04 (2023.01); G06N 20/00 (2019.01); G16H 10/60 (2018.01); G16H 15/00 (2018.01); G16H 50/20 (2018.01)
CPC G06N 5/04 (2013.01) [G06N 20/00 (2019.01); G16H 10/60 (2018.01); G16H 15/00 (2018.01); G16H 50/20 (2018.01)] 8 Claims
OG exemplary drawing
 
1. A method implemented by a computer system, the method comprising:
receiving data representing medical records of a patient, the patient having been diagnosed with diabetes mellitus;
generating a model input using the data representing the medical records of the patient;
determining a first predicted rate of hypoglycemic events by processing the model input using a first machine learning model that has been trained using first training data that comprises data representing medical records of a first plurality of patients and corresponding rates of hypoglycemic events for the respective patients, wherein each of the first plurality of patients uses a first type of insulin;
determining a second predicted rate of hypoglycemic events by processing the model input using a second machine learning model that has been trained using second training data that comprises data representing medical records of a second plurality of patients and corresponding rates of hypoglycemic events for the respective patients, wherein each of the second plurality of patients uses a second type of insulin, the second type of insulin being different from the first type of insulin, and the second plurality of patients being different from the first plurality of patients;
comparing the first predicted rate to the second predicted rate; and
recommending an insulin treatment for the patient based on the comparing.