A Deep Learning Framework Using Enhanced Convolutional Neural Network for Detection of Lung Cancer from CT Images

Lung cancer is one kind of cancer which is causing deaths at an alarming rate across the globe. For patients to recover, early identification and treatment are essential. Histopathological images of tissue biopsies from possibly infected lung regions are used by medical practitioners to make diagnoses. The majority of the time, lung cancer cases are difficult to diagnose and take a long time. Convolutional neural networks are essential for figuring out the best course of therapy for patients and their chance of survival since they can quickly and accurately recognize and categorize different forms of lung cancer. In this paper, A deep learning framework is proposed for automatic detection of lung cancer. The convolutional Neural Network (CNN) model is enhanced for better diagnosis of lung cancer. An algorithm known as a learning-based Method for Lung Cancer Detection (LbM-LCD) is proposed to realize our framework. The empirical study is made with the LUNA-16 dataset. Our experimental results showed that the enhanced CNN outperforms the baseline CNN model with 98.34% accuracy.