Early Detection of Lung Cancer using Hybrid Histological Image Analysis with XGBoost and LightGBM in MATLAB

Lung cancer remains one of the deadliest diseases globally due to the lack of precise and early diagnostic methods. This study proposes a hybrid AI-driven framework for lung cancer detection using histological images, integrating deep learning-based and handcrafted feature extraction with ensemble machine learning models, XGBoost and LightGBM. The framework enhances image quality through noise reduction and contrast adjustment before extracting relevant features for classification. By combining predictions from both classifiers, the model improves accuracy, sensitivity, and specificity compared to traditional methods. Experimental results demonstrate the effectiveness of this approach in automating histopathological analysis, enabling faster and more reliable lung cancer detection, which can significantly enhance patient survival rates.