Lung cancer nodules are a major cause of cancer-related deaths globally. It is essential for the prompt diagnosis of lung cancer that these nodules are found early on chest CT images. However, the heterogeneity and morphological complexity of 2-D nodule features often lead to low detection sensitivity and high false-positive rates, making accurate detection challenging. To tackle these concerns, a detection system aided by CT image has been created to enhance the sensitivity of detection and accuracy of classification of nodules in the lungs. The present plan suggests an efficient algorithm for predicting lung cancer nodules, which is based on an enhanced RCNN. The objective is to address the shortcomings of the existing detection techniques that include inadequate precision and sluggish pace. The suggested approach entails pre-processing the initial CT copy and subsequently forwarding it to the RCNN for the identification of pulmonary nodules The identification outcomes are subsequently employed to finalize the categorization of harmless and cancerous lung growths via the Region-centric Convolutional Neural Network. The findings acquired from the LUNA16 dataset experiment indicate that the enhanced network architecture can achieve a mAP score of 96.71% and a detection rate of 41.99 FPS. These findings imply that the suggested network has the potential to offer an efficient diagnostic instrument for possible nodules of lung cancer and could hold encouraging prospects in the field of clinical practice.