Predictive Modelling For Lung Cancer Detection Using Machine Learning Techniques

Abstract

Early and reliable identification of lung cancer plays a critical role in improving patient survival. This paper presents a computer‑aided diagnosis framework for the detection of lung cancer from computed tomography (CT) images using digital image processing and classical machine learning techniques. The proposed pipeline consists of image preprocessing, lesion segmentation, feature extraction and supervised classification. Noise reduction and contrast enhancement are applied during preprocessing, followed by edge‑based and watershed segmentation to isolate candidate regions. Both region‑based and texture‑based features are extracted from the segmented images and are used to train Support Vector Machine (SVM), Random Forest (RF) and Artificial Neural Network (ANN) classifiers. Performance is evaluated using accuracy, precision, recall and F1‑score. Experimental results indicate that the ANN classifier achieves the best overall performance for both feature categories. The proposed approach demonstrates that a structured feature‑driven machine learning framework can effectively support automated lung cancer detection from CT images.