Machine learning model for rapidly predicting cell segmentation of spatial transcriptomic cell data
WO2025166162A1PCT designated stage Publication Date: 2025-08-07FRED HUTCHINSON CANCER CENT
Patent Information
- Application Number
- PCT/US2025/014029
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-07-09
- Filing Date
- 2025-01-31
- Publication Date
- 2025-08-07
Smart Images
Figure US2025014029_07082025_PF_FP_ABST
Abstract
Biological landmark segmentation may comprise disambiguating which spatial transcriptomics data is associated with background noise and / or partitioning distinct biological landmark types, biological landmarks, and / or biological landmark features. An iterative process may be used to determine such a segmentation, the iterative process comprising modeling gene expression rates according to a current state of the segmentation; randomly altering a current state of the segmentation; determining an updated model of the gene expression rate based on the alteration; and determining to accept or reject the alteration based on a likelihood determined based on the updated model.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Method and apparatus for analysis of chromatin interaction data
US20190295684A1
Genomic safe harbors for genetic therapies in human stem cells and engineered nanoparticles to provide targeted genetic therapies
US20210171983A1
Methods and systems for determining gene expression profiles and cell identities from multi-omic imaging data
US20220180975A1
Cited By
Umbilical cord stem cell three-dimensional segmentation method
CN121281050A