Pulmonary Pathology Detection: An AI-Based Approach with Convolutional Neural Networks

• Pablo Ramirez A. UAI

DOI: https://doi.org/10.5753/cibse.2025.35303

Abstract

Early diagnosis of lung diseases significantly improves survival. This study uses an action research approach with CNNs, ResNet152, and Vision Transformers, optimized with data augmentation. Applying the ISO 25059:2023 standard, the study addresses data interpretability, robustness, and security. Initial results show an accuracy of over 93%, supported by heat maps that highlight key areas in the images, facilitating clinical work. Ethical principles are integrated to protect patient privacy. The project seeks to reduce diagnostic errors and promote the safe use of artificial intelligence in hospitals, with a view to future clinical validation.

Keywords: Convolutional Neural Networks (CNNs), ResNet152, Vision Transformers, Data Augmentation, ISO 25059:2023 Standard, Artificial Intelligence

References

Cai, W., Wang, X., Zhang, X., & Zhang, Y. (2023). Explainable AI for Medical Imaging: Techniques and Applications. Journal of Medical Imaging and Health Informatics, 13(2), 345-358.

Chen, L., Wu, H., Xu, Y., & Zhou, Z. (2022). Deep Learning Approaches for Lung Nodule Detection: Current Trends and Future Directions. International Journal of Computer-Assisted Radiology and Surgery, 17(4), 567-580.

Huang, J., Li, F., Zhao, M., & Yang, T. (2021). Data Augmentation Techniques for Medical Image Analysis: A Comprehensive Review. Computerized Medical Imaging and Graphics, 92, 101967.

Liu, Y., Lu, H., Zhang, J., & Sun, H. (2023). Advancements in Vision Transformers for Medical Image Analysis: Challenges and Opportunities. Artificial Intelligence in Medicine, 135, 102525.

Martínez, R., Sánchez, A., & Paredes, P. (2020). Enhancing Lung Cancer Detection with Hybrid Deep Learning Models: Combining CNNs and Transformers. Journal of Thoracic Imaging, 35(3), 210-219.

Tang, S., Zhou, L., & Chen, P. (2021). Automated Lung Cancer Diagnosis Using Deep Learning Techniques: A Review of Recent Advances. Computational and Structural Biotechnology Journal, 19, 2453-2467.

Wang, J., Zhao, L., & Li, R. (2022). Improving Model Interpretability in Medical AI: The Role of Explainability Tools. Frontiers in Medicine, 9, 858342.

Ramirez Amador, P. (2023). Algoritmos de Inteligencia Artificial para la Detección de Patologías Relacionadas con el Cáncer de Pulmón a través del Análisis de Imágenes utilizando Redes Neuronales Convolucionales y Data Augmentation: un mapeo sistemático de la literatura.

Amador, P. R., Ortega, D. M., & Cesarano, A. (2024). Detection of pulmonary pathologies using convolutional neural networks, Data Augmentation, ResNet50 and Vision Transformers. arXiv preprint arXiv:2409.14446.