基于CycleGAN从DRR合成二维X线透视影像用于肺癌图像引导放疗
本研究开发并评估了一种改进的CycleGAN深度神经网络,用于在肺癌放疗中将数字重建影像DRR转换为平板探测器二维透视影像。模型在超过400对配对的DRR-2D-FPD图像上训练,并在100张二维FPD图像上测试。结果显示合成图像与真实图像高度相似,在MAE、PSNR、SSIM和KID指标上均优于DRR输入和基于U-Net的方法,单张生成时间平均为毫秒级。
要点
- 合成二维FPD图像与真实图像高度相似,可用于放疗摆位验证参考
- 模型可生成逼真的噪声纹理,可能减少手动噪声调整步骤
- 单张图像生成时间仅毫秒级,具备在线应用潜力
核心结论
- 该DNN可将胸部病例的DRR高效转换为逼真的二维FPD图像,有望改善患者摆位验证和临床工作流程。
- 模型生成的噪声模式与真实图像高度相似,且性能优于传统方法。
研究设计与样本
- 本研究为方法学开发与评估研究,使用改进的CycleGAN,在超过400对肺癌患者配对的DRR-2D-FPD图像上训练,在100张二维FPD图像上测试。
- 评估指标包括平均绝对误差MAE、峰值信噪比PSNR、结构相似性指数SSIM和Kernel Inception Distance KID,并记录生成时间。
关键影像表现/征象
- 合成二维FPD图像与真实图像高度相似,仍存在部分位置配准不匹配。
- 定性评估显示模型可生成逼真的噪声模式。
诊断标准与关键数值
- 该DNN在MAE、PSNR、SSIM和KID四个指标上均优于DRR输入和基于U-Net的方法。
- 图像生成平均时间为每张毫秒级。
临床应用与阅片要点
- 该技术可用于肺癌放疗中的二维摆位验证,加速图像引导流程。
- 合成图像的逼真噪声特性可能减少对手动噪声调整的依赖。
局限性
- 合成图像存在部分位置配准不匹配。
- 未来需要在不同成像系统上验证该方法,并解决标记物可视化问题以促进更广泛的临床应用。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVE: We developed and evaluated a deep neural network (DNN) to generate flat-panel detector (FPD) images from digitally reconstructed radiography (DRR) in lung cancer treatment, aiming to streamline image-guided radiotherapy workflows.
METHODS: A modified CycleGAN was trained on over 400 paired DRR-2D-FPD images from lung tumor patients and tested on 100 2D-FPD images. Mean absolute error (MAE), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and Kernel Inception Distance (KID) assessed the synthetic images' similarity to ground truth. Generation time was also recorded.
RESULTS: Despite some positional mismatches, synthetic 2D-FPD images closely resembled ground-truth images. The DNN outperformed both DRR inputs and a U-Net-based method in MAE, PSNR, SSIM, and KID. Image generation averaged milliseconds per image. Qualitatively, the DNN replicated realistic noise patterns, potentially reducing the need for manual noise adjustments.
CONCLUSIONS: This DNN efficiently converts DRR to realistic 2D-FPD images for thoracic cases, potentially enhancing patient setup verification and overall clinical workflow. Future work should validate this approach across various imaging systems and address marker visualization challenges for broader clinical adoption.
ADVANCES IN KNOWLEDGE: This novel CycleGAN-based method rapidly produces realistic 2D-FPD images from DRR, generating noise patterns that closely resemble those of real images and outperforming traditional methods, and may thereby help streamline clinical workflows.
引用
- Hayashi, Mori, Mori, et al. Development of an Unpaired Deep Neural Network for Synthesizing X-ray Fluoroscopic Images from Digitally Reconstructed Tomography in Image Guided Radiotherapy. Br J Radiol. 2026. PMID: 42765334. doi:10.1093/bjr/tqag227
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。