Automatic Detection of Lung Cancer Using Hybrid Classifiers with Hilbert Transform and Detrend Fluctuation Analysis (DFA) Through Chest CT Dataset

Automated identification of lung cancer uses computer algorithms, especially analysing medical images with deep learning and image processing detects cancerous patterns in CT scans and X-rays efficiently. It is presented by the research that Hybrid Classifier Approach for Automatic Lung Cancer Detection Using Chest CT Data (ALCHDC) is a hybrid framework which combines deep learning and signal processing to achieve precise lung cancer classification from Chest CT and X-ray images. The features are combined and purified with Linear Discriminant Analysis (LDA) and a Modified Gravitational Search Algorithm (MGSA), thus enhancing their discriminative ability to a greater extent. The extracted features are used to train a hybrid Convolutional Neural Network (CNN) - Long Short-Term Memory (LSTM) model that not only identifies the spatial patterns but also the sequential dependencies. The evaluation metrics is histogram for hilbert transform, loss, Alex Net accuracy and loss, specificity, recall and f1-score calculation for Alex Net, and comparative analysis.