Multi-channel Residual Neural Network Based on Squeeze-and-Excitation for Osteoporosis Diagnosis

Chunmei Xia,Yue Ding,Jionglin Wu,Wenqiang Luo,Peidong Guo,Tianfu Wang,Baiying Lei
DOI: https://doi.org/10.1007/978-3-031-23179-7_2
2023-01-01
Abstract:Osteoporosis is a progressive, systemic skeletal disease, which is likely to occur in postmenopausal women. The osteoporosis detection utilizing bone mineral density (BMD) measurements by the dualenergy x-ray absorptiometry (DXA) device is expensive and highly ionizing. Bone quantitative ultrasound (QUS) has been regarded as a potential alternative for osteoporosis screening and diagnosis. However, the diagnosis accuracy of QUS is quite low using speed of sound (SOS). Currently, the deep learning method has shown powerful feature extraction capabilities from medical data. In order to improve the diagnosis accuracy of osteoporosis, we propose a multi-channel residual neural network via squeeze-and-excitation attention mechanism (MAResNet), which can extract discriminative features from radio-frequency (RF) signals generated from QUS. Compared with the conventional QUS method using SOS, experimental results indicate that our proposed method achieves superior performance, which can be beneficial to the osteoporosis screening.
What problem does this paper attempt to address?