US 12,142,352 B2
Modular patient analytics
Bilal Bashir, Methuen, MA (US); and Trevor Merrill, Londonderry, MA (US)
Assigned to KONINKLIJKE PHILIPS N.V., Eindhoven (NL)
Filed by KONINKLIJKE PHILIPS N.V., Eindhoven (NL)
Filed on Oct. 4, 2019, as Appl. No. 16/592,834.
Claims priority of provisional application 62/749,706, filed on Oct. 24, 2018.
Prior Publication US 2020/0168305 A1, May 28, 2020
Int. Cl. G16H 10/60 (2018.01); G06F 3/0482 (2013.01); G06F 3/04847 (2022.01); G06N 20/00 (2019.01)
CPC G16H 10/60 (2018.01) [G06N 20/00 (2019.01); G06F 3/0482 (2013.01); G06F 3/04847 (2013.01)] 25 Claims
OG exemplary drawing
 
1. A modular patient analytics system, comprising:
a database comprising site data, the site data comprising patient data and site training data, wherein the patient data comprises one or more types;
an integrated machine learning module configured to: (i) receive the patient data extracted from the site data and (ii) apply one or more machine learning models to the received patient data;
a model training service configured to: (i) receive the site training data extracted from the site data, and (ii) train a first machine learning model using the site training data to generate a first trained machine learning model;
machine learning services that includes the one or more machine learning models comprising: (i) one or more default machine learning models, wherein the one or more default machine learning models are trained using external patient data, other than the site data; and (ii) one or more trained machine learning models, wherein the one or more trained machine learning models comprises the first trained machine learning model;
a patient analytics modules graphical user interface configured to allow a user to select which of the one or more machine learning models to use from the machine learning services, wherein the patient analytics modules graphical user interface is configured to present a list of the one or more machine learning models based upon the one or more types of the patient data in the site data, and is further configured to receive, by the user, the selection of which of the one or more machine learning models to use from the machine learning services; and
a testing service configured to: (i) receive real time patient data from an electronic medical records system, (ii) run, using the received real time patient data, the selected one or more of the one or more default machine learning models or the one or more trained machine learning models from the machine learning services to produce a model output, and (iii) provide the model output to the integrated machine learning module.