A robust framework for shoulder implant X-ray image classification
| DOI | https://doi.org/10.1108/DTA-08-2021-0210 |
| Published date | 30 November 2021 |
| Date | 30 November 2021 |
| Pages | 447-460 |
| Subject Matter | Library & information science,Librarianship/library management,Library technology,Information behaviour & retrieval,Metadata,Information & knowledge management,Information & communications technology,Internet |
| Author | Minh Thanh Vo,Anh H. Vo,Tuong Le |
A robust framework for shoulder
implant X-ray image classification
Minh Thanh Vo
Faculty of Information Technology, HUTECH University,
Ho Chi Minh City, Vietnam
Anh H. Vo
Natural Language Processing and Knowledge Discovery Laboratory,
Faculty of Information Technology, Ton Duc Thang University,
Ho Chi Minh City, Vietnam, and
Tuong Le
Informetrics Research Group, Ton Duc Thang University, Ho Chi Minh City, Vietnam
Abstract
Purpose –Medical images are increasingly popular; therefore, the analysis of these images based on deep
learning helps diagnose diseases become more and more essential and necessary. Recently, the shoulder
implant X-ray image classification(SIXIC) datasetthat includes X-ray images of implanted shoulder prostheses
produced by four manufacturers was released. The implant’s model detection helps to select the correct
equipment and procedures in the upcoming surgery.
Design/methodology/approach –This study proposes a robust model named X-Net to improve the
predictability for shoulder implants X-ray image classification in the SIXIC dataset. The X-Net model utilizes
the Squeeze and Excitation (SE) block integrated into Residual Network (ResNet) module. The SE module aims
to weigh each feature map extracted from ResNet, which aids in improving the performance. The feature
extraction process of X-Net model is performed by both modules: ResNet and SE modules. The final feature is
obtained by incorporating the extracted features from the above steps, which brings more important
characteristics of X-ray images in the input dataset. Next, X-Net uses this fine-grained feature to classify the
input images into four classes (Cofield, Depuy, Zimmer and Tornier) in the SIXIC dataset.
Findings –Experiments are conducted to show the proposed approach’s effectiveness compared with other
state-of-the-art methods for SIXIC. The experimental results indicate that the approach outperforms the
various experimental methods in terms of several performance metrics. In addition, the proposed approach
provides the new state of the art results in all performance metrics, such as accuracy, precision, recall, F1-score
and area under the curve (AUC), for the experimental dataset.
Originality/value –The proposed method with high predictive performance can be used to assist in the
treatment of injured shoulder joints.
Keywords Shoulder implant X-ray classification, Medical images analysis, Squeeze and excitation,
Residual network, Fine-tuning approach
Paper type Research paper
1. Introduction
Deep learning currently plays an increasingly important role in computer vision (Vo et al.,
2019,2020;Mao et al., 2021;Quiroz et al., 2020;Kerkech et al., 2020), natural language
processing (Nguyen et al., 2019;Vo et al., 2021) and time series prediction (Le et al., 2019,2020).
In the above problems, deep learning is used first and significantly effective for computer
vision with many various applications, such as robotic (Vo et al., 2020), environmental
research (Vo et al., 2019;Mao et al., 2021), smart agriculture (Quiroz et al., 2020;Kerkech et al.,
2020), medical diagnosis (Santos et al., 2020;Le and Baik, 2019;Pereira et al., 2019;Filho et al.,
2019;Itani et al., 2019) and object detection (Zhenyu et al., in press). Recently, there has been a
lot of research interest in medical diagnosis using machine learning and deep learning. Santos
et al. (2020) developed a convolutional neural network (CNN)-based model to distinguish
between eight types of tumors (i.e. four benign and four malignant). Next, Le and Baik (2019)
presented a framework that uses a sampling technique and extreme gradient boosting for
Medical images
447
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/2514-9288.htm
Received 9 August 2021
Revised 5 October 2021
Accepted 8 November 2021
Data Technologies and
Applications
Vol. 56 No. 3, 2022
pp. 447-460
© Emerald Publishing Limited
2514-9288
DOI 10.1108/DTA-08-2021-0210
To continue reading
Request your trial