Design and Evaluation of a Customized CNN Architecture for Early Detection of Lung Cancer

We have chosen to conduct research on a biological phrase, namely lung cancer detection, taking into account the most recent trends and technological advancements. In recent years, image processing techniques have become widely employed in many medical fields to enhance images at early stages of cancer identification and therapy. There are many different forms of cancer, including those of the lungs, breast, blood, throat, brain and mouth. Lung cancer is a condition in which aberrant cells proliferate and develop into tumors. The key to treating lung cancer is finding it when it is still in its early stages. Medical issues can arise in anyone. One of the most erratic illnesses a person may experience is cancer. In many regions of the world, extensive MRI screening is used is still unfeasible, keeping midsection radiology in the beginning and most fundamental method. It is important to note that due to CNN’s superior performance and capability, these models are better able to identify illnesses like lung cancer. The accuracy gained with Convolutional Neural Network(CNN) is between 80 and 90%, which is superior than the accuracy of more established, conventional techniques. This is accomplished by using CT scan data series for lung cancer to use the convolutional neural network approach.