AI-Assisted Early Screening of Lung Cancer: Evaluation of Deep Learning Models based on CT Images

Liyan Zhong

doi:10.54097/dmwrfd35

Authors

Liyan Zhong

DOI:

https://doi.org/10.54097/dmwrfd35

Keywords:

AI-Assisted Screening, Early Diagnosis of Lung Cancer, CT Image Analysis, Deep Learning, Multi-Scale Attention Fusion Convolutional Neural Network (MSA-FCNN)

Abstract

Aiming at the problems of low efficiency and insufficient accuracy of manual reading in early CT image screening of lung cancer, this paper proposes an original multi-scale attention fusion convolutional neural network (MSA-FCNN). The model extracts multi-level features based on the improved residual network, obtains information of different scales through the multi-scale feature generation module, and combines the spatial and channel attention mechanisms to highlight the key lesion features, and finally achieves benign/malignant binary classification. The experiment was conducted on the LIDC-IDRI dataset and a private dataset. The results showed that the accuracy of MSA-FCNN on the LIDC-IDRI dataset was 96.8%, the sensitivity was 95.2%, the specificity was 97.5%, and the AUC was 0.983; the accuracy was 95.3%, the sensitivity was 93.8%, the specificity was 96.7%, and the AUC was 0.976 on the private dataset, which were significantly better than mainstream algorithms such as ResNet-50. This model provides an efficient and reliable auxiliary tool for early screening of lung cancer.

Downloads

Download data is not yet available.

References

[1] Zhu, Z., Chen, M., Song, L., Wang, J., Hu, G., Han, W., ... & Jin, Z. CT image weighted genomics score predicts the efficacy of immunotherapy for non-small cell lung cancer. Journal of Chinese Academy of Medical Sciences, Vol. 45(2023) No. 5, p. 794-802.

[2] Li, X., Zhang, P. & Duan, L. Research progress of 18F-FDG PET/CT imaging genomics in targeted diagnosis and treatment of non-small cell lung cancer. International Journal of Radiology and Nuclear Medicine, Vol. 49(2025) No. 6, p. 383-388.

[3] Fu, Y., Hou, R. & Fu, X. Research progress on predicting the risk of lymphatic or hematogenous metastasis of early non-small cell lung cancer based on chest CT. Chinese Journal of Oncology, Vol. 32(2022) No. 4, p. 343-350.

[4] Wang, X., Luo, H., Li, K., Yang, J., Jing, Y. & Chen, W. Prediction of distant metastasis of non-small cell lung cancer based on CT radiomics combined prediction model. Oncology, Vol. 31(2022) No. 4, p. 357-366.

[5] Ai, J., Gao, H., Ai, S., Li, H., Shi, G. & Wei, Y. Diagnostic value of CT radiomics for cystic lung cancer. Journal of Shandong University (Medical Edition), Vol. 61(2024) No. 12, p. 70-77.

[6] Han, Z., Chen, Y., Yang, Y., Gong, J., Jia, S. & Nie, S. Research progress of CT radiomics in the evaluation of therapeutic efficacy of non-small cell lung cancer. Oncology, Vol. 43(2023) No. 8, p. 692-700.

[7] Yin, K., Wei, W., Xiong, H., Huang, J. & Dai, H. Clinical and high-resolution CT imaging features and EGFR in patients with stage I~IIIB resectable peripheral non-small cell lung cancer: Correlation study of gene mutation. International Journal of Radiation Medicine and Nuclear Medicine, Vol. 47(2023) No. 8, p. 461-467.

[8] Chen, X. & Yang, B. Application of radiomics technology in the prognosis of non-small cell lung cancer. CT Theory and Application Research, Vol. 33(2024) No. 3, p. 385-390.

[9] Wei, Y., Lu, Z. & Ren, Y. The mechanism of action of Ki-67 in lung cancer and the research progress of radiomics. Journal of China Medical University, Vol. 53(2024) No. 3, p. 271-275.

[10] Wu, C., He, C., Chen, Z., Zhao, D., Zhou, L. & Zhao, L. CT imaging features and differential diagnosis of pulmonary pseudotumor and peripheral lung cancer. Chinese Journal of CT and MRI, Vol. 20(2022) No. 4, p. 51-52.

[11] Li, L., Yang, J. & Chen, Y. Application and progress of PET/CT in evaluating the effect of immunotherapy for non-small cell lung cancer. Oncology Imaging, Vol. 32(2023) No. 2, p. 114-121.

[12] Guo, Y., Zhang, L., Huang, Y., Yao, X., Zeng, M. & Huang, J. Application value of random forest algorithm based on CT plain scan data for classification and identification of pathological types of primary lung cancer. Chinese Journal of Medical Equipment, Vol. 38(2023) No. 2, p. 30-35.

[13] Liu, D., Xu, H., Wang, X., Yang, Z., Li, W., Liu, H. & Xie, Z. The value of CT radiomics in predicting lymph node metastasis of non-small cell lung cancer. Journal of Bengbu Medical University, Vol. 46(2021) No. 9, p. 1239-1243.