Lung cancer disease identification using hybrid models

Abstract

Using hybrid models, we present a novel method for detecting lung cancer in this study. Our method uses the random forest and convolutional neural network (CNN) techniques to incorporate machine learning and deep learning advantages. The proposed composite method combines structured clinical data with unprocessed imaging data for a more complete lung cancer diagnosis. The CNN component of our hybrid model excels at extracting features from images of lung cancer, while the random forest component excels at capturing complex relationships in structured data. For greater precision and consistency, the results of the two models may be averaged. The hybrid model outperforms the existing methods. The hybrid model acquired an accuracy rate of 98%. Future lung cancer detection will be rapid and accurate due to the hybrid model’s improved performance and decreased inference periods.