An Intelligent Detection System for Pulmonary Nodules in Medical CT Images based on Deep Learning with a Voice Prompt System

Xuan Zhang; Dingwang Zhang

doi:10.54097/1meqcj30

Authors

Xuan Zhang
Dingwang Zhang

DOI:

https://doi.org/10.54097/1meqcj30

Keywords:

Pulmonary Nodules, Deep Learning, Voice Prompt, Computer-Aided Diagnosis

Abstract

Lung cancer is one of the malignant tumors with the highest mortality rate globally, and its high mortality is closely related to late-stage diagnosis. Low-dose computed tomography (LDCT) has been confirmed as an effective screening method to reduce lung cancer mortality in high-risk populations, but it significantly increases the workload and diagnostic pressure on radiologists. To address this challenge, this paper proposes and implements an integrated intelligent diagnostic system. The system combines a high-performance, two-stage deep learning model for the automated detection of pulmonary nodules and innovatively integrates a voice prompt module aimed at optimizing the workflow of radiologists and improving human-computer interaction efficiency. The detection model proposed in this paper adopts a 3D context-aware network architecture and includes a dedicated false-positive reduction module to enhance clinical utility. The main contributions of this study are: 1) designing and validating a high-precision pulmonary nodule detection model; and 2) exploring a novel paradigm for AI-assisted diagnostic interaction by introducing a voice prompt system, which aims to improve diagnostic efficiency and accuracy, providing new insights for the clinical translation of intelligent medical image analysis.

Downloads

Download data is not yet available.

References

[1] Ning, J., Ge, T., Jiang, M., Jia, K., Wang, L., Li, W., ... & He, Y. (2021). Early diagnosis of lung cancer: which is the optimal choice?. Aging (Albany NY), 13(4), 6214.

[2] Gao, C., Wu, L., Wu, W., Huang, Y., Wang, X., Sun, Z., ... & Gao, C. (2025). Deep learning in pulmonary nodule detection and segmentation: a systematic review. European radiology, 35(1), 255-266.

[3] Stec, N., Arje, D., Moody, A. R., Krupinski, E. A., & Tyrrell, P. N. (2018). A systematic review of fatigue in radiology: is it a problem?. American Journal of Roentgenology, 210(4), 799-806.

[4] Abdullah, M., & Shaukat, F. (2025). A Unified Multi-Scale Attention-Based Network for Automatic 3D Segmentation of Lung Parenchyma & Nodules In Thoracic CT Images. arXiv preprint arXiv:2505.17602.

[5] Hardie, R. C., Trout, A. T., Dillman, J. R., Narayanan, B. N., & Tanimoto, A. A. (2024). Performance analysis in children of traditional and deep learning CT lung nodule computer-aided detection systems trained on adults. American Journal of Roentgenology, 222(2), e2330345.

[6] RamSoft (June 23, 2025) How AI Is Helping in Radiology Automation and Efficiency? https://www. ramsoft.com/ blog/ radiology-automation.

[7] Khalaji, A., Riahi, F., Rafieezadeh, D., Khademi, F., Fesharaki, S., & Joni, S. S. (2025). Artificial intelligence in automated detection of lung nodules: a narrative review. International Journal of Physiology, Pathophysiology and Pharmacology, 17(2), 45.

[8] Aase, A., Fabbrini, A. E., White, K. M., Averill, S., Gravely, A., & Melzer, A. C. (2020). Implementation of a standardized template for reporting of incidental pulmonary nodules: feasibility, acceptability, and outcomes. Journal of the American College of Radiology, 17(2), 216-223.

[9] Min, J. Y., & Jung, G. (2025). Automating Radiology Report Generation: A Systematic Review of Deep Learning Methods.

[10] Shaip (Jan 8, 2024) The Power of Natural Language Processing (NLP) in Radiology: Enhancing Diagnosis and Efficiency. https://weareshaip.medium.com/the-power-of-natural-language-processing-nlp-in-radiology-enhancing-diagnosis-and-efficiency-83c37fc491f3.

[11] Ma, H., Guo, H., Zhao, M., Qi, S., Li, H., Tian, Y., ... & Qian, W. (2022). Automatic pulmonary ground‐glass opacity nodules detection and classification based on 3D neural network. Medical Physics, 49(4), 2555-2569.

[12] Ni, Y., Xie, Z., Zheng, D., Yang, Y., & Wang, W. (2022). Two-stage multitask U-Net construction for pulmonary nodule segmentation and malignancy risk prediction. Quantitative Imaging in Medicine and Surgery, 12(1), 292.

[13] Zheng, S., Kong, S., Huang, Z., Pan, L., Zeng, T., Zheng, B., ... & Liu, Z. (2022). A lower false positive pulmonary nodule detection approach for early lung cancer screening. Diagnostics, 12(11), 2660.

[14] Srivastava, D., Srivastava, S. K., Khan, S. B., Singh, H. R., Maakar, S. K., Agarwal, A. K., ... & Albalawi, E. (2023). Early detection of lung nodules using a revolutionized deep learning model. Diagnostics, 13(22), 3485.

[15] Nguyen, C. C., Tran, G. S., Burie, J. C., & Nghiem, T. P. (2021). Pulmonary nodule detection based on faster R-CNN with adaptive anchor box. Ieee Access, 9, 154740-154751.

[16] Setio, A. A. A., Traverso, A., De Bel, T., Berens, M. S., Van Den Bogaard, C., Cerello, P., ... & Jacobs, C. (2017). Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge. Medical image analysis, 42, 1-13.

[17] Aase, A., Fabbrini, A. E., White, K. M., Averill, S., Gravely, A., & Melzer, A. C. (2020). Implementation of a standardized template for reporting of incidental pulmonary nodules: feasibility, acceptability, and outcomes. Journal of the American College of Radiology, 17(2), 216-223.

[18] Yu, S., Lee, S. S., & Hwang, H. (2024). The ethics of using artificial intelligence in medical research. Kosin Medical Journal, 39(4), 229-237.

[19] Note: This citation is used for all methods in Table 1 as per the source document.

[20] Zheng, S., Kong, S., Huang, Z., Pan, L., Zeng, T., Zheng, B., ... & Liu, Z. (2022). A lower false positive pulmonary nodule detection approach for early lung cancer screening. Diagnostics, 12(11), 2660.

[21] Deepika, P., Prasanth, T., Ramu, Y., & Tanmayi, R. (2022). False positive reduction of lung nodule detection using deep learning techniques. Int. J. Res. Publ. Rev, 3(11), 321-328.