公开(公告)号：US20220207708A1

公开(公告)日：2022-06-30

申请号：US17571399

申请日：2022-01-07

Applicant: Genentech, Inc.

Inventor： Zheng LI ,  Calvin TSAY

IPC: G06T7/00 ,  G06N20/00 ,  G06N3/04 ,  G06F16/53 ,  G06F17/15

Abstract: In one embodiment, a method includes receiving one or more querying images associated with a container of a pharmaceutical product, each of the one or more querying images being based on a particular angle of the container of the pharmaceutical product, calculating one or more confidence scores associated with one or more defect indications, respectively for the container of the pharmaceutical product, by processing the one or more querying images using a target machine-learning model, and determining a defect indication for the container of the pharmaceutical product from the one or more defect indications based on a comparison between the one or more confidence scores and one or more predefined threshold scores, respectively.

公开(公告)号：US20240020817A1

公开(公告)日：2024-01-18

申请号：US18474076

申请日：2023-09-25

Applicant: Genentech, Inc.

Inventor： Zheng LI ,  Calvin Tsay

IPC: G06T7/00 ,  G06N20/00 ,  G06N3/04 ,  G06F16/53 ,  G06F17/15

CPC classification number: G06T7/0004 ,  G06N20/00 ,  G06N3/04 ,  G06F16/53 ,  G06F17/15 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: In one embodiment, a method includes receiving one or more querying images associated with a container of a pharmaceutical product, each of the one or more querying images being based on a particular angle of the container of the pharmaceutical product, calculating one or more confidence scores associated with one or more defect indications, respectively for the container of the pharmaceutical product, by processing the one or more querying images using a target machine-learning model, and determining a defect indication for the container of the pharmaceutical product from the one or more defect indications based on a comparison between the one or more confidence scores and one or more predefined threshold scores, respectively.

公开(公告)号：US20240084241A1

公开(公告)日：2024-03-14

申请号：US18464135

申请日：2023-09-08

Applicant: GENENTECH, INC.

Inventor： Arthi NARAYANAN ,  Aditya Avdhut WALVEKAR ,  Georo L. ZHOU ,  Nicholas RUMMEL ,  Zheng LI ,  Steven J. MEIER

IPC: C12M1/36 ,  G16B40/20

CPC classification number: C12M41/48 ,  G16B40/20

Abstract: A method, system, and non-transitory computer readable medium for predicting a glycan distribution of one or more glycans attached to molecules during a biomolecules manufacturing process are disclosed. In various embodiments, at least three manufacturing process parameters related to the process for manufacturing the molecules are input into a probabilistic graphical model that is trained to predict glycan distribution. The trained probabilistic graphical model may then analyze the at least three manufacturing process parameters to predict the distribution of the glycans that are attached to the molecules.

公开(公告)号：US20240084240A1

公开(公告)日：2024-03-14

申请号：US18464131

申请日：2023-09-08

Applicant: GENENTECH, INC.

Inventor： Arthi NARAYANAN ,  Aditya Avdhut WALVEKAR ,  Georo L. ZHOU ,  Nicholas RUMMEL ,  Zheng LI ,  Steven J. MEIER

IPC: C12M1/36 ,  G06N3/08

CPC classification number: C12M41/48 ,  G06N3/08

Abstract: A method, system, and non-transitory computer readable medium for predicting cell viability of a cell culture in a bioreactor during a biomolecule manufacturing process are disclosed. In various embodiments, at least three manufacturing process parameters related to the process for manufacturing molecules are input into a machine learning model that is trained to predict cell viabilities. The trained machine learning model may then analyze the at least three manufacturing process parameters to generate an indicator of cell viability of the cell culture.

公开(公告)号：US20220309666A1

公开(公告)日：2022-09-29

申请号：US17841610

申请日：2022-06-15

Applicant: Genentech, Inc.

Inventor： Zheng LI ,  Mandy Man Chu YIM ,  Bibi EPHRAIM ,  Dat TRAN ,  Xinyu LIU ,  David SHAW

IPC: G06T7/00 ,  G06T7/11 ,  G06T7/73 ,  G06V10/25 ,  G06V10/20 ,  G06V10/26 ,  G06V10/46

Abstract: A method includes, by a computing system, receiving a querying image depicting a sampling area, processing the querying image using a single cluster detection model to identify one or more regions of the querying image depicting a cluster in the sampling area, processing the one or more regions using a cluster verification deep-learning model to determine whether each depicted cluster is a cell cluster, and determining that exactly one of the identified one or more regions depicts a cluster that is a cell cluster. The method further includes processing the region depicting the cell cluster using a morphology deep-learning model to determine that there is only one cell in the cell cluster and to determine that die morphology of the cell is acceptable.