目的 提出一种新型的肋骨骨折检测网络Rib-Net，探讨其进行肋骨骨折检测的可行性与准确性，以减少骨折漏诊案例。  方法 采用公开数据集RibFrac Dataset，其数据集划分为训练集（420例）、验证集（80例）及测试集（160例）。Rib-Net由目标检测集成网络ED-Net、完全候选框融合算法（CBF）与分割模型3D Unet构成。首先，集成Retina Unet、UFRCNN+与Mask RCNN组成ED-Net，预测肋骨骨折候选框；其次，设计全新的CBF，融合存在重叠的骨折候选框，生成定位精准、置信度准确的候选框；最后，利用Unet对肋骨骨折进行分割，实现肋骨骨折的进一步精确定位。  结果 在“MICCAI 2020 RibFrac Challenge: Rib Fracture Detection and Classification”挑战赛平台上，Rib-Net检测结果达到了最优成绩，其召回率、无限制接受者操作特性曲线（FROC）值及Dice相似指数分别为92.3%，0.859和0.61。  结论 Rib-Net网络可高效精准地对胸部CT影像进行肋骨骨折检测定位，有效协助医生做出准确诊断。

Objective  To propose a new rib fracture detection network Rib-Net to automatically and accurately detect and locate rib fracture and address the issue of missed diagnosis of rib fractures.   Methods  The public data set RibFrac Dataset was used to evaluate the performance of the Rib-Net, and the data set was divided into training set (420 cases), validation set (80 cases), and test set (160 cases). The Rib-Net was composed of the object detection integrated network Ensemble Detection Net (ED-Ne), Complete Box Fusion (CBF) module and the segmentation network 3D Unet. Firstly, Retina Unet, UFRCNN+ and Mask RCNN were integrated to form ED-Net to predict rib fracture candidate boxes. Secondly, a new CBF module was designed to fuse overlapping fracture candidate boxes to generate candidate boxes with accurate positioning and accurate confidence. Finally, Unet was used for rib fracture segmentation to achieve further precise localization of rib fractures.   Results  On the “MICCAI 2020 RibFrac Challenge: Rib Fracture Detection and Classification challenge”, our proposed Rib-Net’s detection results  reached the best performance, and its recall rate, free-response  receiver operating characteristic curve(FROC) value and Dice were 92.3%, 0.859 and 0.61, respectively.   Conclusion  The Rib-Net network can efficiently and accurately detect and locate rib fractures on chest CT images, effectively assisting doctors in making accurate diagnosis.