TY - JOUR
T1 - Integrating Transfer Learning and Feature Aggregation into Self-defined Convolutional Neural Network for Automated Detection of Lung Cancer Bone Metastasis
AU - Guo, Yanru
AU - Lin, Qiang
AU - Wang, Yubo
AU - Cao, Xu
AU - Cao, Yongchun
AU - Man, Zhengxing
AU - Zeng, Xianwu
AU - Huang, Xiaodi
N1 - Funding Information:
This work was supported by the Key R&D Plan of Gansu Province (Grant No. 21YF5GA063), the Natural Science Foundation of Gansu Province (Grant No. 20JR5RA511), the Youth Ph.D. Foundation of Education Department of Gansu Province (Grant No. 2021QB-063), the Fundamental Research Funds for the Central Universities (Grant Nos. 31920220020, 31920210054, 31920210013), the National Natural Science Foundation of China (Grant No. 61562075), the Gansu Provincial First-class Discipline Program of Northwest Minzu University (Grant No. 11080305), and the Program for Innovative Research Team of SEAC (Grant No. [2018] 98).
Publisher Copyright:
© 2022, Taiwanese Society of Biomedical Engineering.
PY - 2023/2
Y1 - 2023/2
N2 - Purpose: To improve the diagnosis accuracy and efficiency of lung cancer bone metastasis routinely performed by nuclear medicine physicians, we propose a deep learning-based image classification model that can learn the features from two views of an image first, then aggregate them, and finally classify the image into the presence or absence of bone metastasis. Methods: We present a new network that can automatically classify scintigraphy images collected from the clinical diagnosis of metastasis in patients with lung cancer. The proposed network consists of pre-training, transfer learning, and two-view feature aggregation. In the pre-training stage, the proposed model is trained on a source dataset of Chest X-Ray. In the transfer learning stage, the pre-trained model is fine-turned on the target dataset of scintigraphy images. The extracted features from anterior and posterior views of an image are aggregated in the final stage. The classification network can detect the presence or absence of metastases in scintigraphy images. Results: Experimental evaluations on a set of clinical scintigraphy images showed that the proposed network performed well for automatically classifying metastatic images with the mean scores of 0.7710, 0.8311, 0.6827 and 0.7475 on the accuracy, precision, recall and F-1 score, respectively. Conclusion: The proposed classification network can predict whether an image shows lung cancer-caused metastasis with state-of-the-art performance.
AB - Purpose: To improve the diagnosis accuracy and efficiency of lung cancer bone metastasis routinely performed by nuclear medicine physicians, we propose a deep learning-based image classification model that can learn the features from two views of an image first, then aggregate them, and finally classify the image into the presence or absence of bone metastasis. Methods: We present a new network that can automatically classify scintigraphy images collected from the clinical diagnosis of metastasis in patients with lung cancer. The proposed network consists of pre-training, transfer learning, and two-view feature aggregation. In the pre-training stage, the proposed model is trained on a source dataset of Chest X-Ray. In the transfer learning stage, the pre-trained model is fine-turned on the target dataset of scintigraphy images. The extracted features from anterior and posterior views of an image are aggregated in the final stage. The classification network can detect the presence or absence of metastases in scintigraphy images. Results: Experimental evaluations on a set of clinical scintigraphy images showed that the proposed network performed well for automatically classifying metastatic images with the mean scores of 0.7710, 0.8311, 0.6827 and 0.7475 on the accuracy, precision, recall and F-1 score, respectively. Conclusion: The proposed classification network can predict whether an image shows lung cancer-caused metastasis with state-of-the-art performance.
KW - Bone scan
KW - Convolutional neural network
KW - Image classification
KW - Lung cancer
KW - Metastasis
UR - http://www.scopus.com/inward/record.url?scp=85144652085&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85144652085&partnerID=8YFLogxK
U2 - 10.1007/s40846-022-00770-z
DO - 10.1007/s40846-022-00770-z
M3 - Article
SN - 2199-4757
VL - 43
SP - 53
EP - 62
JO - Journal of Medical and Biological Engineering
JF - Journal of Medical and Biological Engineering
IS - 1
ER -