Department of Radiology, The Affiliated Jiangning Hospital of Nanjing Medical University, No. 168, gushan Road, Nanjing, 211100, Jiangsu Province, China.
Institute of Advanced Research, Beijing Infervision Technology Co Ltd, Yuanyang International Center, Beijing, 100025, China.
Eur Radiol. 2021 Jun;31(6):3815-3825. doi: 10.1007/s00330-020-07418-z. Epub 2020 Nov 17.
To develop a convolutional neural network (CNN) model for the automatic detection and classification of rib fractures in actual clinical practice based on cross-modal data (clinical information and CT images).
In this retrospective study, CT images and clinical information (age, sex and medical history) from 1020 participants were collected and divided into a single-centre training set (n = 760; age: 55.8 ± 13.4 years; men: 500), a single-centre testing set (n = 134; age: 53.1 ± 14.3 years; men: 90), and two independent multicentre testing sets from two different hospitals (n = 62, age: 57.97 ± 11.88, men: 41; n = 64, age: 57.40 ± 13.36, men: 35). A Faster Region-based CNN (Faster R-CNN) model was applied to integrate CT images and clinical information. Then, a result merging technique was used to convert 2D inferences into 3D lesion results. The diagnostic performance was assessed on the basis of the receiver operating characteristic (ROC) curve, free-response ROC (fROC) curve, precision, recall (sensitivity), F1-score, and diagnosis time. The classification performance was evaluated in terms of the area under the ROC curve (AUC), sensitivity, and specificity.
The CNN model showed improved performance on fresh, healing, and old fractures and yielded good classification performance for all three categories when both clinical information and CT images were used compared to the use of CT images alone. Compared with experienced radiologists, the CNN model achieved higher sensitivity (mean sensitivity: 0.95 > 0.77, 0.89 > 0.61 and 0.80 > 0.55), comparable precision (mean precision: 0.91 > 0.87, 0.84 > 0.77, and 0.95 > 0.70), and a shorter diagnosis time (average reduction of 126.15 s).
A CNN model combining CT images and clinical information can automatically detect and classify rib fractures with good performance and feasibility in actual clinical practice.
• The developed convolutional neural network (CNN) performed better in fresh, healing, and old fractures and yielded a good classification performance in three categories, if both (clinical information and CT images) were used compared to CT images alone. • The CNN model had a higher sensitivity and matched precision in three categories than experienced radiologists with a shorter diagnosis time in actual clinical practice.
基于跨模态数据（临床信息和 CT 图像），开发一种用于在实际临床实践中自动检测和分类肋骨骨折的卷积神经网络（CNN）模型。
在这项回顾性研究中，收集了 1020 名参与者的 CT 图像和临床信息（年龄、性别和病史），并将其分为单中心训练集（n=760；年龄：55.8±13.4 岁；男性：500）、单中心测试集（n=134；年龄：53.1±14.3 岁；男性：90）和来自两家不同医院的两个独立多中心测试集（n=62，年龄：57.97±11.88，男性：41；n=64，年龄：57.40±13.36，男性：35）。应用 Faster Region-based CNN（Faster R-CNN）模型来整合 CT 图像和临床信息。然后，使用结果合并技术将 2D 推断转换为 3D 病变结果。根据受试者工作特征（ROC）曲线、自由响应 ROC（fROC）曲线、精度、召回率（敏感性）、F1 评分和诊断时间评估诊断性能。使用 ROC 曲线下面积（AUC）、敏感性和特异性来评估分类性能。
CNN 模型在新鲜、愈合和陈旧性骨折方面表现出了更好的性能，并且与仅使用 CT 图像相比，当同时使用临床信息和 CT 图像时，该模型对所有三种类型的骨折都具有良好的分类性能。与有经验的放射科医生相比，CNN 模型具有更高的敏感性（平均敏感性：0.95>0.77、0.89>0.61 和 0.80>0.55）、可比的精度（平均精度：0.91>0.87、0.84>0.77 和 0.95>0.70）和更短的诊断时间（平均减少 126.15 秒）。
一种结合 CT 图像和临床信息的 CNN 模型可以在实际临床实践中自动检测和分类肋骨骨折，具有良好的性能和可行性。
与仅使用 CT 图像相比，开发的卷积神经网络（CNN）在新鲜、愈合和陈旧性骨折方面表现更好，如果同时使用（临床信息和 CT 图像），则在三个类别中都具有更好的性能和可行性。
在实际临床实践中，与有经验的放射科医生相比，CNN 模型在三个类别中具有更高的敏感性和匹配精度，且诊断时间更短。
