Breast Lesion Ultrasound Video Dataset
收藏资源简介:
本研究首次收集并标注了一个用于乳腺病变检测的超声视频数据集,包含188个视频,旨在支持深度学习模型在视频级别的乳腺病变检测。数据集由厦门大学信息学院计算机科学系创建,涵盖了从肿瘤出现到最大部分再到消失的完整扫描过程。数据集中的视频由LOGIQ-E9和PHILIPS TIS L9-3设备获取,由两位具有八年乳腺病理学经验的病理学家手动标注病变区域并分类。该数据集主要用于训练和评估乳腺病变检测模型,特别是在动态视频中准确识别病变的能力。
This study is the first to collect and annotate an ultrasound video dataset for breast lesion detection, which contains 188 videos and is intended to support deep learning models in conducting video-level breast lesion detection. The dataset was developed by the Department of Computer Science, School of Information Science and Engineering, Xiamen University, and covers the complete scanning process from the appearance of breast lesions, through the stage of their largest size, to their final disappearance. The videos in the dataset were acquired using LOGIQ-E9 and PHILIPS TIS L9-3 ultrasound devices, and the lesion regions were manually annotated and classified by two pathologists with eight years of experience in breast pathology. This dataset is primarily used for training and evaluating breast lesion detection models, especially their capacity to accurately identify lesions in dynamic videos.

- 1A New Dataset and A Baseline Model for Breast Lesion Detection in Ultrasound Videos厦门大学信息学院计算机科学系 · 2022年



