Lung cancer detection and classification using medical imaging plays a pivotal role in early diagnosis and treatment planning. In this study, an Optimized Hard Voting Classifier (OHVC) is proposed to detect and classify lung cancer from CT images, categorizing them into three distinct classes: Benign, Malignant, and Normal. The approach combines the strengths of both deep learning and machine learning models to improve classification accuracy and robustness. Specifically, a combination of Convolutional Neural Networks (CNNs), Hybrid Convolutional Recurrent Neural Networks (HNNs), Support Vector Machines (SVMs), Naive Bayes (NBs), Random Forests (RFs), Deep Neural Networks (DNNs), and Recurrent Neural Networks (RNNs) are leveraged. These models are trained on preprocessed CT images, with each classifier contributing its individual prediction to a majority voting mechanism for the final class determination. The ensemble approach ensures improved accuracy of 0.99, reducing the risk of overfitting while handling the inherent variability of lung imaging data. Experimental results demonstrate the effectiveness of this method, achieving high classification performance in distinguishing between benign, malignant, and normal lung tissue, making it a promising tool for clinical decision support in lung cancer diagnosis.