| CPC G06T 7/0012 (2013.01) [G06V 20/698 (2022.01); G06T 2207/10056 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30024 (2013.01)] | 34 Claims |

|
1. A computer-implemented method comprising:
a) receiving cell imaging data comprising at least one training transmitted light micrograph;
b) generating a cell imaging dataset comprising time course data of transmitted light micrographs comprising cells having a cell state representing at least one phase of a cell differentiation process collected at an interval over a time period, wherein the transmitted light micrographs comprise at least one of a brightfield micrograph, a phase contrast micrograph, or a differential interference contrast (DIC) micrograph;
c) applying a machine learning model configured to analyze the cell imaging dataset to generate mathematical representations of the transmitted light micrographs, build a plurality of profiles predictive of at least one cell state, and generate a trajectory of transition of the cells from the at least one cell state to another cell state; and
d) determining the presence of one or more cells having the at least one cell state and how far the one or more cells are along the trajectory in an experimental transmitted light micrograph based on the plurality of profiles and mathematical representations of the experimental transmitted light micrograph.
|