Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network Model
This study proposes a novel multi-network architecture consisting of a multi-scale convolution neural network (MSCNN) with fully connected graph convolution network (GCN), named MSCNN-GCN, for the detection of musculoskeletal abnormalities via musculoskeletal radiographs. To obtain both detailed and...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/20/11/3153 |
id |
doaj-6e1eaf6da6a046ebae8eb2baf86f3fcc |
---|---|
record_format |
Article |
spelling |
doaj-6e1eaf6da6a046ebae8eb2baf86f3fcc2020-11-25T02:57:24ZengMDPI AGSensors1424-82202020-06-01203153315310.3390/s20113153Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network ModelShuang Liang0Yu Gu1School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing 100083, ChinaSchool of AutoMation, Guangdong University of Petrochemical Technology, Maoming 525000, ChinaThis study proposes a novel multi-network architecture consisting of a multi-scale convolution neural network (MSCNN) with fully connected graph convolution network (GCN), named MSCNN-GCN, for the detection of musculoskeletal abnormalities via musculoskeletal radiographs. To obtain both detailed and contextual information for a better description of the characteristics of the radiographs, the designed MSCNN contains three subnetwork sequences (three different scales). It maintains high resolution in each sub-network, while fusing features with different resolutions. A GCN structure was employed to demonstrate global structure information of the images. Furthermore, both the outputs of MSCNN and GCN were fused through the concat of the two feature vectors from them, thus making the novel framework more discriminative. The effectiveness of this model was verified by comparing the performance of radiologists and three popular CNN models (DenseNet169, CapsNet, and MSCNN) with three evaluation metrics (Accuracy, F1 score, and Kappa score) using the MURA dataset (a large dataset of bone X-rays). Experimental results showed that the proposed framework not only reached the highest accuracy, but also demonstrated top scores on both F1 metric and kappa metric. This indicates that the proposed model achieves high accuracy and strong robustness in musculoskeletal radiographs, which presents strong potential for a feasible scheme with intelligent medical cases.https://www.mdpi.com/1424-8220/20/11/3153abnormality detectionCNNfusionGCNmulti-networkmusculoskeletal radiographs |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Shuang Liang Yu Gu |
spellingShingle |
Shuang Liang Yu Gu Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network Model Sensors abnormality detection CNN fusion GCN multi-network musculoskeletal radiographs |
author_facet |
Shuang Liang Yu Gu |
author_sort |
Shuang Liang |
title |
Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network Model |
title_short |
Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network Model |
title_full |
Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network Model |
title_fullStr |
Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network Model |
title_full_unstemmed |
Towards Robust and Accurate Detection of Abnormalities in Musculoskeletal Radiographs with a Multi-Network Model |
title_sort |
towards robust and accurate detection of abnormalities in musculoskeletal radiographs with a multi-network model |
publisher |
MDPI AG |
series |
Sensors |
issn |
1424-8220 |
publishDate |
2020-06-01 |
description |
This study proposes a novel multi-network architecture consisting of a multi-scale convolution neural network (MSCNN) with fully connected graph convolution network (GCN), named MSCNN-GCN, for the detection of musculoskeletal abnormalities via musculoskeletal radiographs. To obtain both detailed and contextual information for a better description of the characteristics of the radiographs, the designed MSCNN contains three subnetwork sequences (three different scales). It maintains high resolution in each sub-network, while fusing features with different resolutions. A GCN structure was employed to demonstrate global structure information of the images. Furthermore, both the outputs of MSCNN and GCN were fused through the concat of the two feature vectors from them, thus making the novel framework more discriminative. The effectiveness of this model was verified by comparing the performance of radiologists and three popular CNN models (DenseNet169, CapsNet, and MSCNN) with three evaluation metrics (Accuracy, F1 score, and Kappa score) using the MURA dataset (a large dataset of bone X-rays). Experimental results showed that the proposed framework not only reached the highest accuracy, but also demonstrated top scores on both F1 metric and kappa metric. This indicates that the proposed model achieves high accuracy and strong robustness in musculoskeletal radiographs, which presents strong potential for a feasible scheme with intelligent medical cases. |
topic |
abnormality detection CNN fusion GCN multi-network musculoskeletal radiographs |
url |
https://www.mdpi.com/1424-8220/20/11/3153 |
work_keys_str_mv |
AT shuangliang towardsrobustandaccuratedetectionofabnormalitiesinmusculoskeletalradiographswithamultinetworkmodel AT yugu towardsrobustandaccuratedetectionofabnormalitiesinmusculoskeletalradiographswithamultinetworkmodel |
_version_ |
1724711490606333952 |