Search academic texts

Information × Registration Number 0123U103488, R & D request Title Highly parallel implementation of Local Binary Pattern algorithm for high-throughput analysis imaging pipeline in spatial transcriptomics (fastLBP) Head Frolova Alina O., Registration Date 18-08-2023 Organization Institute of Molecular Biology and Genetics of NAS of Ukraine popup.description1 We plan to implement a highly parallel version of the visual descriptor used for classification in computer vision, namely the Local Binary Pattern (LBP) algorithm. This algorithm has a minimal number of input parameters, thus, can serve as a baseline method for many biological studies, while having the potential to process hundreds of high-resolution images in a short time. However, current implementations of LBP need to be significantly improved in terms of parallelization. We will use LBP to establish a high throughput analysis pipeline to quantify tissue architecture from H&E image data in spatial transcriptomics. We will apply LBP to map the regional architecture of glioblastoma (GBM) brain tumors from large collections of tumor H&E images. popup.nrat_date 2024-12-09 Close