Lung Cancer Detection and Classification Using Efficient Net

Abstract

Early prediction and classification of cancer stages are mandated to take countermeasures for treatment and easy diagnosis. The scanned images are mostly used to obtain the occurrence of lung small cell cancer. To diagnose lung cancer using EfficientNet we present a convolutional neural network to diagnose three types of lung cancer based on scanned images. The proposed model consists of the  main path and three sub-paths. The main path works to extract the small features and creates feature maps at low-level. As for the sub-paths is responsible for transferring the medium and high levels feature maps to fully connected layers to complete the classification process, also the ResNet was prepared to compare it with the performance of the proposed. We use neural architecture search to design a new baseline network and scale it up to obtain a family of models, called EfficientNets. We systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance. Based on this observation, we propose a new scaling method that uniformly scales all dimensions of depth/width/resolution using a simple yet highly effective compound coefficient.