GIRAFE
收藏资源简介:
GIRAFE数据集是由布雷斯特大学和格雷戈里奥·马拉尼奥医院合作创建的,旨在促进对声带高速视频内窥镜序列的高级语义分割、分析和评估。该数据集包含65个高速视频记录,来自50名患者(30名女性,20名男性),涵盖了从健康到各种声带疾病的广泛情况。所有记录均由专家手动标注,并结合了传统和深度学习方法的自动分割。数据集的创建过程包括数据收集、标注和验证,旨在支持声带疾病诊断和新算法开发,特别是在声带间隙的自动分割方面。
The GIRAFE Dataset was co-developed by the University of Brest and Gregorio Marañón Hospital, aiming to facilitate advanced semantic segmentation, analysis and evaluation of high-speed videoendoscopic sequences of the vocal folds. This dataset comprises 65 high-speed video recordings from 50 patients (30 female, 20 male), covering a broad spectrum of conditions ranging from healthy vocal folds to various vocal fold disorders. All recordings were manually annotated by experts, paired with automatic segmentations derived via both traditional and deep learning methodologies. The dataset development process includes data collection, annotation and validation, and is designed to support vocal fold disease diagnosis and the development of novel algorithms, particularly for automatic segmentation of the glottal space.

- 1GIRAFE: Glottal Imaging Dataset for Advanced Segmentation, Analysis, and Facilitative Playbacks Evaluation信息医学处理实验室(LaTIM),UMR 1101,INSERM 布雷斯特大学 · 2024年



