Affiliation:
1. School of Computer Science and Engineering, IFTM University, Moradabad, India
2. Faculty of Engineering and Technology, M.J.P. Rohilkhand University, Bareilly, India
Abstract
Cancer stands out as a disease with a high mortality rate around the world, and some types of cancer, namely lung, skin, nervous system, breast, colorectal, prostate area, and blood-related cancers, are more threatening. The final goal of the chapter is to help improve early cancer diagnosis and come up with possible ways to fight the terrible effects of these deadly diseases. Lung cancer is a very dangerous disease that can grow quickly and spread to other parts of the body through a process called dissemination. It is essential to detect and precisely diagnose cancerous lung nodules to begin therapy as soon as possible. CT scan pictures are being sorted by using machine learning, especially advanced CNN models like vgg16 ang ResNet50v2. There are a variety of obstacles, such as abnormalities in nodule patterns, shapes, and sizes, as well as contextual complications. The goal of this study is to improve precision by using preliminary processing, methods like ADASYN, and solving class imbalance.
Reference26 articles.
1. Abunajm, S., Elsayed, N., ElSayed, Z., & Ozer, M. (2023). Deep Learning Approach for Early Stage Lung Cancer Detection. arXiv preprint arXiv:2302.02456.
2. Transfer learning with GoogLeNet for detection of lung cancer
3. Al-Yasriy, H. F., Al-Husieny, M. S., Mohsen, F. Y., Khalil, E. A., & Hassan, Z. S. (2020, November). Diagnosis of lung cancer based on CT scans using CNN. In IOP Conference Series: Materials Science and Engineering (Vol. 928, No. 2, p. 022035). IOP Publishing. https://iopscience.iop.org/article/10.1088/1757-899X/928/2/022035/meta
4. Comparison of deep learning convolutional neural network (CNN) architectures for CT lung cancer classification