US 12,482,556 B2
Value-based advanced clinical benchmarking
Luca Neri, Milan (IT); Francesco Bellocchio, Siziano (IT); Carlo Barbieri, Crema (IT); Stefano Stuard, Bad Homburg (DE); Jasmine Ion Titapiccolo, Cernusco sul Naviglio (IT); and Paola Carioni, Credera Rubbiano (IT)
Assigned to Fresenius Medical Care Deutschland GmbH, Bad Homburg (DE)
Filed by Fresenius Medical Care Deutschland GmbH, Bad Homburg (DE)
Filed on May 13, 2022, as Appl. No. 17/743,751.
Claims priority of application No. 21180742 (EP), filed on Jun. 22, 2021.
Prior Publication US 2022/0406444 A1, Dec. 22, 2022
Int. Cl. G16H 40/20 (2018.01); G16H 40/63 (2018.01); G16H 50/20 (2018.01); G16H 50/70 (2018.01)
CPC G16H 40/20 (2018.01) [G16H 40/63 (2018.01); G16H 50/20 (2018.01); G16H 50/70 (2018.01)] 18 Claims
OG exemplary drawing
 
1. A system, comprising:
a plurality of dialysis machines;
a plurality of gateways in communication with the plurality of dialysis machines; and
one or more servers;
wherein the plurality of dialysis machines are configured to:
perform dialysis treatments for respective patients associated with the plurality of dialysis machines;
obtain data associated with the respective patients, including treatment-related data; and
send the obtained data to the one more servers via the plurality of gateways and a communication network;
wherein the one or more servers are configured to:
receive the data associated with the respective patients;
compute, based on the received data, for the respective patients, expected frequencies of at least one type of health outcome, wherein computing the expected frequencies of the at least one type of health outcome comprises: obtaining one or more analytical models for a set of features extracted from the data and for the at least one type of health outcome; and
generalize the computed expected frequencies of the at least one type of health outcome for the respective patients into one or more individual-level impact metrics associated with risk factors for the at least one type of health outcome, wherein the risk factors comprise one or more modifiable risk factors and one or more un-modifiable risk factors; and
wherein the one or more servers are further configured to:
set up a model architecture based on a Continuous Quality Improvements (CQI) structure; and
train n+m risk models based on the obtained one or more analytical models and the model architecture, wherein n represents the number of the at least one type of health outcome and m represents the number of the one or more modifiable risk factors.