Lung cancer (LC) is one of the chief causes of the death globally and hence, it is listed in the top 5 most occurring cancers. Appropriate diagnosis and treatment is necessary for treatment planning and implementation. The clinical level screening is commonly performed using the Computed-Tomography (CT) and the images generated using this scheme is examined to confirm the harshness of the LC. In this research, a deep-learning scheme is proposed to classify the lung CT into normal and cancer class. The different phases of this scheme includes; image collection and resizing, feature extraction using the DenseNet scheme, feature reduction with 50% dropout and serial concatenation, and bi-level classification using the chosen classifiers. The analysis confirms that the suggested approach offers a detection accuracy of > 96% with the Support Vector Machine (SVM) classifier. The investigation is carried out using Python software. This study demonstrates that the suggested plan can be taken into consideration in order to look at the LC from the selected lung CT database.