Automatic detection and classification of peri-prosthetic femur fracture.
Automatic detection and classification of peri-prosthetic femur fracture.
复制标题
DOI：
10.1007/s11548-021-02552-5
复制
发表时间：
2022-04
影响因子：
3
通讯作者：
Xie SQ
中科院分区：
文献类型：
--
作者：
Alzaid A;Wignall A;Dogramadzi S;Pandit H;Xie SQ
目标分类和定位是计算机辅助诊断(CAD)工具的一项关键任务。虽然已经有许多为CAD开发的通用深度学习(DL)模型，但在文献中还没有工作来评估它们在诊断关节植入物附近的骨折时的有效性。在这项工作中，我们的目的是评估现有的分类系统在二类和多类问题(骨折类型)上的性能，使用普通的射线照片。此外，我们还评估了采用单级和两级DL结构的目标检测系统的性能。收集了1272张股骨假体周围骨折的X线片。两名临床专家用包围盒注释骨折，并根据温哥华分类系统(A、B、C型)对骨折进行分类。评价了Densenet161、Resnet50、Inception、VGG四种分类模型和FASTER RCNN、RetinanNet两种目标检测模型，并比较了它们的性能。报告了六种基于混淆矩阵的方法来评估骨折分类。对于骨折的定位，报告了平均精度和定位精度。Resnet50在骨折/正常的二进制分类中表现出最好的准确性和F1评分。此外，Resnet50在多分类(正常、温哥华A、B和C型)中显示出准确性。通过两个独立的评估创建了PFF图像的大数据集和骨折特征的注释，以实现基于DL的PFF检测、分类和定位方法。结果表明，该方法可作为关节植入物附近骨折的一种有前途的诊断工具。
Object classification and localization is a key task of computer-aided diagnosis (CAD) tool. Although there have been numerous generic deep learning (DL) models developed for CAD, there is no work in the literature to evaluate their effectiveness when utilized in diagnosing fractures in proximity of joint implants. In this work, we aim to assess the performance of existing classification systems on binary and multi-class problems (fracture types) using plain radiographs. In addition, we evaluated the performance of object detection systems using the one- and two-stage DL architectures. A data set of 1272 X-ray images of Peri-prosthetic Femur Fracture PFF was collected. The fractures were annotated with bounding boxes and classified according to the Vancouver Classification System (type A, B, C) by two clinical specialists. Four classification models such as Densenet161, Resnet50, Inception, VGG and two object detection models such as Faster RCNN and RetinaNet were evaluated, and their performance compared. Six confusion matrix-based measures were reported to evaluate fracture classification. For localization of the fracture, Average Precision and localization accuracy were reported. The Resnet50 showed the best performance with accuracy and F1-score in the binary classification: fracture/normal. In addition, the Resnet50 showed accuracy in multi-classification (normal, Vancouver type A, B and C). A large data set of PFF images and the annotations of fracture features by two independent assessments were created to implement a DL-based approach for detecting, classifying and localizing PFFs. It was shown that this approach could be a promising diagnostic tool of fractures in proximity of joint implants.
登录
查看更多内容
DOI：
10.1073/pnas.1806905115
发表时间：
2018-11-06
影响因子：
11.1
作者：
Lindsey R;Daluiski A;Chopra S;Lachapelle A;Mozer M;Sicular S;Hanel D;Gardner M;Gupta A;Hotchkiss R;Potter H
通讯作者：
Potter H
影响因子：
4.4
作者：
Kitamura, Gene;Chung, Chul Y.;Moore, Barry E., II
通讯作者：
Moore, Barry E., II
影响因子：
12
作者：
Joshi, Deepa;Singh, Thipendra P.
通讯作者：
Singh, Thipendra P.
DOI：
10.1007/s11548-020-02150-x
发表时间：
2020-04-25
影响因子：
3
作者：
Jimenez-Sanchez, Amelia;Kazi, Anees;Mateus, Diana
通讯作者：
Mateus, Diana
影响因子：
5.9
作者：
Cheng, Chi-Tung;Ho, Tsung-Ying;Liao, Chien-Hung
通讯作者：
Liao, Chien-Hung