Chenyang Hong

TL;DR: A new class of sequence patterns that flexibly model such variable regions, and corresponding algorithms that identify co-bound sequences using these patterns are proposed and shown to lead to better classification performance than patterns that do not explicitly model the variable regions.

TL;DR: This work addresses the problem of automatic nuclear segmentation and classification within the Haematoxylin and Eosin stained histology images with a simultaneous semantic and instance segmentation framework and demonstrates, through visual and numerical experimental, that the model outperforms the provided baselines by a large margin.

TL;DR: In this article , the authors proposed the COMSV method that is specifically designed for cancer samples and applied it to cancer cell lines and patient samples to identify hundreds of novel structural variations per sample.