DLCP: A Robust Deep Learning with Non-linear CA Mechanism for Lung Cancer Prediction

Lung cancer is identified as the most dangerous hazards faced by both female and male. Lung cancer may result various breathing problems during inhaling and exhaling. Many static and dynamic methods are invented to predict this disease and mitigate the effect. The death rate due to lung cancer is increasing in old and young people which is alarming. We have studied the existing literature and found that, there is a need for a comprehensive, adoptable and accurate versatile classifier for the lung cancer to predict the same at an early stage to mitigate the mortality rate. We have collected 1, 37, 896 datasets (CT images) from UCI machine learning repository to train and test our classifier. We have applied non-linear cellular automata augmented with convolution neural network (CNN) for predicting the lung cancer. The proposed classifier DLCP (Deep Learning Cancer Prediction) is compared with standard baseline methods with the parameters accuracy, error rate, specificity, sensitivity, ROC and precision. DLCP reports an average accuracy of 98.49% which is promising compared with the cited literature.