US 12,475,976 B2
Population PK/PD linking parameter analysis using deep learning
James Lu, San Francisco, CA (US)
Assigned to Genentech, Inc., South San Francisco, CA (US)
Filed by Genentech, Inc., South San Francisco, CA (US)
Filed on Sep. 23, 2022, as Appl. No. 17/951,781.
Application 17/951,781 is a continuation of application No. PCT/US2021/024257, filed on Mar. 25, 2021.
Claims priority of provisional application 62/994,701, filed on Mar. 25, 2020.
Prior Publication US 2023/0018216 A1, Jan. 19, 2023
Int. Cl. G16C 20/70 (2019.01); G16C 20/50 (2019.01)
CPC G16C 20/70 (2019.02) [G16C 20/50 (2019.02)] 18 Claims
OG exemplary drawing
 
1. A method for predicting a set of linking parameters that relate pharmacokinetic and pharmacodynamic effects, the method comprising:
receiving, by one or more processors, a population dataset that comprises a population pharmacokinetic (PK) dataset and a population pharmacodynamic (PD) dataset;
transforming, by the one or more processors, the population dataset into a plurality of data density images that includes a PK data density image and a PD data density image, wherein the PK data density image reflects densities of data points corresponding to selected dose effects and selected points in time, and wherein the PD data density image reflects densities of data points corresponding to selected drug effects and selected points in time; and
predicting, by the one or more processors, the set of linking parameters using the plurality of data density images and a deep learning system, wherein the deep learning system takes the plurality of data density images as input and outputs the predicted set of linking parameters, wherein the deep learning system is trained using a simulated PK dataset and a simulated PD dataset, and wherein the simulated PK dataset is generated based on a simulation of drug concentration with respect to points in time and the simulated PD dataset is generated based on a simulation of drug effect with respect to points in time.