Anatomy-Informed Cephalometric X-ray Generation (AICG) Pipeline
收藏资源简介:
本研究提出了一种创新的数据生成方法,能够生成多样化的头颅侧位X光图像及其对应的标注,无需人工干预。该数据集包括真实头颅侧位X光图像和详细的医疗文本描述,旨在提高头颅侧位标志点检测的准确性和可靠性。该数据集的创建过程包括构建新的头颅侧位标志点标注,并使用扩散模型生成与标注相对应的X光图像。该数据集的应用领域是头颅侧位标志点检测,旨在解决数据稀缺和标注困难的问题。
This study proposes an innovative data generation method that can generate diverse lateral cephalometric X-ray images and their corresponding annotations without manual intervention. This dataset includes real lateral cephalometric X-ray images and detailed medical text descriptions, aiming to improve the accuracy and reliability of lateral cephalometric landmark detection. The creation process of this dataset involves constructing new lateral cephalometric landmark annotations and using diffusion models to generate X-ray images matching the annotations. The application field of this dataset is lateral cephalometric landmark detection, which is designed to solve the problems of data scarcity and annotation difficulties.




