LiTS17(Liver Tumor Segmentation Challenge 2017)

肝脏是原发性（即起源于肝脏，如肝细胞癌，HCC）或继发性（即扩散到肝脏，如结肠直肠癌）肿瘤发展的常见部位。由于它们的异质和扩散形状，肿瘤病变的自动分割非常具有挑战性。到目前为止，只有交互式方法才能获得可接受的分割肝脏病变的结果。
面对我们的挑战，我们鼓励研究人员开发自动分割算法，以在对比增强腹部 CT 扫描中分割肝脏病变。数据和细分由世界各地的各种临床站点提供。训练数据集包含 130 个 CT 扫描，测试数据集包含 70 个 CT 扫描。该挑战赛是与 ISBI 2017 和 MICCAI 2017 一起组织的。对于 MICCAI 2017，我们增加了肝脏分割和肿瘤负荷估计的任务。

The liver is a common site for the development of primary tumors (originating in the liver itself, e.g., hepatocellular carcinoma, HCC) and secondary tumors (spreading to the liver from other sites, e.g., colorectal cancer). Due to their heterogeneous and diffuse shapes, automatic segmentation of tumor lesions is highly challenging. To date, only interactive approaches have achieved acceptable performance for liver lesion segmentation.

In response to this challenge, we encourage researchers to develop automatic segmentation algorithms for liver lesions in contrast-enhanced abdominal CT scans. Data and annotations were provided by various clinical sites across the globe. The training dataset comprises 130 CT scans, while the test dataset includes 70 CT scans. This challenge was organized in conjunction with ISBI 2017 and MICCAI 2017. For MICCAI 2017, we added two additional tasks: liver segmentation and tumor burden estimation.

搜集汇总

背景与挑战

背景概述

LiTS17是一个用于肝脏肿瘤自动分割的医疗影像数据集，包含130个训练CT扫描和70个测试CT扫描，旨在解决肝脏病变异质和扩散形状带来的分割挑战。该数据集与ISBI 2017和MICCAI 2017会议相关，并扩展了肝脏分割和肿瘤负荷估计任务，适用于医学图像分割研究。