Department of Radiology, Seoul National University Hospital, Seoul, Republic of Korea.
Department of Radiology, Seoul National University College of Medicine, Seoul, Republic of Korea.
Pediatr Radiol. 2024 Sep;54(10):1692-1703. doi: 10.1007/s00247-024-05999-1. Epub 2024 Jul 24.
Artificial intelligence has been increasingly used in medical imaging and has demonstrated expert level performance in image classification tasks.
To develop a fully automatic approach for determining the Risser stage using deep learning on abdominal radiographs.
In this multicenter study, 1,681 supine abdominal radiographs (age range, 9-18 years, 50% female) obtained between January 2019 and April 2022 were collected retrospectively from three medical institutions and graded manually using the United States Risser staging system. A total of 1,577 images from Hospitals 1 and 2 were used for development, and 104 images from Hospital 3 for external validation. From each radiograph, right and left iliac crest patch images were extracted using the pelvic bone segmentation model DeepLabv3 + with the EfficientNet-B0 encoder trained with 90 digitally reconstructed radiographs from pelvic computed tomography scans with a pelvic bone mask. Using these patch images, ConvNeXt-B was trained to grade according to the Risser classification. The model's performance was evaluated using accuracy, area under the receiver operating characteristic curve (AUROC), and mean absolute error.
The fully automatic Risser stage assessment model showed an accuracy of 0.87 and 0.75, mean absolute error of 0.13 and 0.26, and AUROC of 0.99 and 0.95 on internal and external test sets, respectively.
We developed a deep learning-based, fully automatic segmentation and classification model for Risser stage assessment using abdominal radiographs.
人工智能已越来越多地应用于医学影像学领域，并在图像分类任务中展现出了专家级的性能。
开发一种基于深度学习的全自动化方法，用于通过腹部 X 光片确定 Risser 分期。
本多中心研究回顾性收集了 2019 年 1 月至 2022 年 4 月期间来自 3 家医疗机构的 1681 例仰卧位腹部 X 光片（年龄范围为 9-18 岁，女性占 50%），并使用美国 Risser 分期系统对其进行人工分级。来自医院 1 和 2 的 1577 张图像用于开发，医院 3 的 104 张图像用于外部验证。从每张 X 光片中，使用骨盆骨骼分割模型 DeepLabv3+提取右侧和左侧髂嵴贴片图像，该模型使用 90 张骨盆计算机断层扫描的数字重建射线照片和骨盆射线照片的骨盆蒙版训练有 EfficientNet-B0 编码器。使用这些贴片图像，根据 Risser 分类训练 ConvNeXt-B 进行分级。使用准确性、受试者工作特征曲线下面积（AUROC）和平均绝对误差来评估模型的性能。
全自动 Risser 分期评估模型在内部和外部测试集上的准确性分别为 0.87 和 0.75，平均绝对误差分别为 0.13 和 0.26，AUROC 分别为 0.99 和 0.95。
我们开发了一种基于深度学习的全自动分割和分类模型，用于通过腹部 X 光片评估 Risser 分期。
