Abstract：
In this work, we applied and validated an artificial intelligence technique known as generative adversarial networks (GANs) to create large volumes of high-fidelity synthetic anteroposterior (AP) pelvis radiographs that can enable deep learning (DL)-based image analyses, while ensuring patient privacy.
AP pelvis radiographs with native hips were gathered from an institutional registry between 1998 and 2018. The data was used to train a model to create 512 × 512 pixel synthetic AP pelvis images. The network was trained on 25 million images produced through augmentation. A set of 100 random images (50/50 real/synthetic) was evaluated by 3 orthopaedic surgeons and 2 radiologists to discern real versus synthetic images. Two models (joint localization and segmentation) were trained using synthetic images and tested on real images.
The final model was trained on 37,640 real radiographs (16,782 patients). In a computer assessment of image fidelity, the final model achieved an \"excellent\" rating. In a blinded review of paired images (1 real, 1 synthetic), orthopaedic surgeon reviewers were unable to correctly identify which image was synthetic (accuracy = 55%, Kappa = 0.11), highlighting synthetic image fidelity. The synthetic and real images showed equivalent performance when they were assessed by established DL models.
This work shows the ability to use a DL technique to generate a large volume of high-fidelity synthetic pelvis images not discernible from real imaging by computers or experts. These images can be used for cross-institutional sharing and model pretraining, further advancing the performance of DL models without risk to patient data safety.
Level III.
AP pelvis radiographs with native hips were gathered from an institutional registry between 1998 and 2018. The data was used to train a model to create 512 × 512 pixel synthetic AP pelvis images. The network was trained on 25 million images produced through augmentation. A set of 100 random images (50/50 real/synthetic) was evaluated by 3 orthopaedic surgeons and 2 radiologists to discern real versus synthetic images. Two models (joint localization and segmentation) were trained using synthetic images and tested on real images.
The final model was trained on 37,640 real radiographs (16,782 patients). In a computer assessment of image fidelity, the final model achieved an \"excellent\" rating. In a blinded review of paired images (1 real, 1 synthetic), orthopaedic surgeon reviewers were unable to correctly identify which image was synthetic (accuracy = 55%, Kappa = 0.11), highlighting synthetic image fidelity. The synthetic and real images showed equivalent performance when they were assessed by established DL models.
This work shows the ability to use a DL technique to generate a large volume of high-fidelity synthetic pelvis images not discernible from real imaging by computers or experts. These images can be used for cross-institutional sharing and model pretraining, further advancing the performance of DL models without risk to patient data safety.
Level III.
摘要：
背景：在这项工作中，我们应用并验证了一种称为生成对抗网络（GAN）的人工智能（AI）技术，以创建大量高保真合成前后（AP）骨盆X射线照片，可以实现基于深度学习（DL）的图像分析。确保患者隐私。
方法：从1998年至2018年的机构注册中收集了具有天然臀部的AP骨盆X射线照片。数据用于训练模型以创建512×512像素的合成AP骨盆图像。该网络接受了通过增强生成的2500万张图像的培训。由三名整形外科医生和两名放射科医生评估了一组100张随机图像（50/50真实/合成），以区分真实图像与合成图像。使用合成图像训练了两个模型（联合定位和分割），并在真实图像上进行了测试。
结果：最终模型在37,640张真实X射线照片（16,782名患者）上进行了训练。在图像保真度的计算机评估中，最终模型获得了“优秀”评级。在对配对图像的盲目审查中(1真实，1合成)，整形外科医生评审人员无法正确识别哪幅图像是合成的(准确率=55%，Kappa=0.11),突出显示合成图像保真度。当通过建立的DL模型评估合成图像和真实图像时，它们显示出等效的性能。
结论：这项工作显示了使用DL技术生成大量高保真合成骨盆图像的能力，这些图像无法从计算机或专家的真实成像中辨别。这些图像可用于跨机构共享和模型预培训，进一步推进DL模型的性能，而不会对患者数据安全造成风险。
方法：从1998年至2018年的机构注册中收集了具有天然臀部的AP骨盆X射线照片。数据用于训练模型以创建512×512像素的合成AP骨盆图像。该网络接受了通过增强生成的2500万张图像的培训。由三名整形外科医生和两名放射科医生评估了一组100张随机图像（50/50真实/合成），以区分真实图像与合成图像。使用合成图像训练了两个模型（联合定位和分割），并在真实图像上进行了测试。
结果：最终模型在37,640张真实X射线照片（16,782名患者）上进行了训练。在图像保真度的计算机评估中，最终模型获得了“优秀”评级。在对配对图像的盲目审查中(1真实，1合成)，整形外科医生评审人员无法正确识别哪幅图像是合成的(准确率=55%，Kappa=0.11),突出显示合成图像保真度。当通过建立的DL模型评估合成图像和真实图像时，它们显示出等效的性能。
结论：这项工作显示了使用DL技术生成大量高保真合成骨盆图像的能力，这些图像无法从计算机或专家的真实成像中辨别。这些图像可用于跨机构共享和模型预培训，进一步推进DL模型的性能，而不会对患者数据安全造成风险。
