BriefGPT.xyz
Oct, 2023
通过视觉和文本提示增强皮肤镜病变分割的扩散模型
Boosting Dermatoscopic Lesion Segmentation via Diffusion Models with Visual and Textual Prompts
HTML
PDF
Shiyi Du, Xiaosong Wang, Yongyi Lu, Yuyin Zhou, Shaoting Zhang...
TL;DR
应用扩散模型结合病灶特定的视觉和文本提示生成皮肤镜检测图像,相对于传统生成模型在图像质量和皮损分割性能上表现更优,SSIM图像质量指标提高9%,Dice系数超过先前方法5%以上。
Abstract
image synthesis
approaches, e.g.,
generative adversarial networks
, have been popular as a form of data augmentation in
medical image analysis
→