Review of Deep Learning Techniques for Lung Cancer Detection
摘要
One of the most challenging and elusive forms of cancer is lung cancer. Because nodules typically result in mortality in both men and women, it is critical to check them as soon as possible and accurately. As a result, numerous techniques have been created to identify lung cancer at an early stage. This study compares various early detection techniques based on machine learning. Several methods have been developed in recent years for identifying lung cancer. Though some also use x-rays, CT scans are employed by the majority of them. Image recognition is used to find lung tumor nodules using a variety of segmentation and classifier methods. This study discovered that the majority of the CT scan images used for cancer diagnosis are better suited for giving accurate results. The results of this method are also more accurate when compared to marker-controlled watershed segmentation. The results of deep learning techniques were more accurate than those obtained using conventional machine-learning techniques.