US 12,424,306 B2
Tree-based data exploration and data-driven protocol
Chen Liu, Danvers, MA (US); and Ahmad El Katerji, Danvers, MA (US)
Assigned to ABIOMED, INC., Danvers, MA (US)
Filed by ABIOMED, Inc., Danvers, MA (US)
Filed on Mar. 12, 2024, as Appl. No. 18/602,147.
Application 18/602,147 is a continuation of application No. 17/859,407, filed on Jul. 7, 2022, granted, now 11,955,214.
Application 17/859,407 is a continuation of application No. 16/593,555, filed on Oct. 4, 2019, granted, now 11,420,039, issued on Aug. 23, 2022.
Claims priority of provisional application 62/741,985, filed on Oct. 5, 2018.
Prior Publication US 2024/0371487 A1, Nov. 7, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. A61B 5/00 (2006.01); A61M 60/113 (2021.01); A61M 60/17 (2021.01); A61M 60/178 (2021.01); A61M 60/205 (2021.01); A61M 60/216 (2021.01); A61M 60/295 (2021.01); A61M 60/50 (2021.01); A61M 60/515 (2021.01); A61M 60/538 (2021.01); A61M 60/592 (2021.01); G06N 20/20 (2019.01); G16H 20/00 (2018.01)
CPC G16H 20/00 (2018.01) [A61B 5/7264 (2013.01); A61B 5/7275 (2013.01); A61M 60/113 (2021.01); A61M 60/17 (2021.01); A61M 60/178 (2021.01); A61M 60/205 (2021.01); A61M 60/216 (2021.01); A61M 60/295 (2021.01); A61M 60/50 (2021.01); A61M 60/515 (2021.01); A61M 60/538 (2021.01); A61M 60/592 (2021.01); G06N 20/20 (2019.01); A61M 2205/04 (2013.01); A61M 2205/3303 (2013.01); A61M 2205/3331 (2013.01); A61M 2205/50 (2013.01); A61M 2210/125 (2013.01); A61M 2230/30 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining training data from a data repository, wherein the training data comprises patient data and operational parameters of a plurality of ventricular assist devices (VADs), and wherein the data repository comprises data from treatment of acute myocardial infarction (AMI) patients, high-risk percutaneous coronary interventions (PCI) patients, or patients in cardiogenic shock;
training a prediction model with the training data, wherein the trained prediction model is configured to use a tree-based machine learning algorithm to predict a plurality of survival rates, and wherein each one of the predicted survival rates is associated with using a different one of the VADs to treat a patient; and
adjusting a hyper-parameter of the tree-based machine learning algorithm to improve both a cross-validation score and a training score.