HVAngleEst: A Dataset for End-to-end Automated Hallux Valgus Angle Measurement from X-Ray Images

Accurate measurement of hallux valgus angle (HVA) and intermetatarsal angle (IMA) is essential for diagnosing hallux valgus and determining appropriate treatment. Traditional manual methods, while standardized, are time-consuming, labor-intensive, and subject to evaluator bias. Recent advancements in deep learning have been applied to hallux valgus angle estimation, but the development of effective algorithms requires large, well-annotated datasets. Existing X-ray datasets are typically limited to cropped foot regions images, and only one dataset containing very few samples is publicly available. To address these challenges, we introduce HVAngleEst, the first large-scale, open-access dataset specifically designed for hallux valgus angle estimation. HVAngleEst comprises 1,382 X-ray images from 1,150 patients and includes comprehensive annotations, such as foot localization, hallux valgus angles, and line segments for each phalanx. This dataset enables fully automated, end-to-end hallux valgus angle estimation, reducing manual labor and eliminating evaluator bias.

搜集汇总

背景与挑战

背景概述

HVAngleEst是首个专为拇外翻角估计设计的大规模开放数据集，包含1,382张X射线图像和来自1,150名患者的全面注释，支持全自动、端到端的角度测量，以克服传统手动方法的局限性和偏差。