A local-global unified scheme driven by positionable texture and multi-level boundary for lung cancer organoids segmentation

Jiansong Fan, Tianxu Lv, Shunyuan Jia, Yuan Liu, Ruihong Deng, Zexin Chen, Yu Zhu, Lihua Li, Chunjuan Jiang, Jianming Ni, Xiang Pan

Abstract: Highlights•PTMBNet: A Novel unified scheme for high-throughput organoid segmentation.•Proposed TFM and TPM for detailed organoid texture feature amid elevated noise.•MBFM and BRM extract multi-level boundaries, ensuring consistency with BTC.•Introducing a large-scale LCOs dataset. PTMBNet surpasses state-of-the-art methods.