TL;DR本研究提出了一种名为多形态感知网络(MMA-Net)的新框架,通过整合多种脊柱形态作为关注信息,改进了 Cobb 角度测量的准确性,实现了自动化的、高效的、可靠的脊柱 X 光分析。
Abstract
scoliosis diagnosis and assessment depend largely on the measurement of the
Cobb angle in spine X-ray images. With the emergence of deep learning
techniques that employ landmark detection, tilt prediction, and sp