• T.KALA

Lung cancer is a major health concern, and its early detection is crucial for improving patient survival rates. However, identifying malignant nodules in computed tomography (CT) scans remains challenging due to the complexity of tumor patterns and class imbalance in medical datasets. This study proposes an advanced deep learning approach using the Progressive Growing Channel Attentive Non-Local (ProCAN) network, combined with the Synthetic Minority Over-sampling Technique (SMOTE) to enhance detection accuracy. ProCAN leverages progressive feature learning, channel attention mechanisms, and non-local dependencies, allowing the model to focus on critical tumor regions while capturing both local and global patterns. SMOTE addresses the imbalance issue by generating synthetic samples for underrepresented classes, ensuring a more robust training process. Experimental results demonstrate improved classification accuracy, enhanced sensitivity to malignant cases, and a reduced false-negative rate. This research highlights the effectiveness of ProCAN with SMOTE in lung cancer detection, offering potential benefits for early diagnosis and treatment planning

Lung cancer detection, Deep learning, Progressive Growing Channel Attentive Non-Local network (ProCAN), Synthetic Minority Over-sampling Technique (SMOTE), Lung Segmentation, Class imbalance, Computed Tomography scan analysis, Channel attention mechanism.

"Lung Cancer Detection using ProCAN with Enhanced Data Balancing Techniques", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 5, page no.p728-p735, May 2023, Available :http://www.jetir.org/papers/JETIR2305G93.pdf

"Lung Cancer Detection using ProCAN with Enhanced Data Balancing Techniques", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 5, page no. ppp728-p735, May 2023, Available at : http://www.jetir.org/papers/JETIR2305G93.pdf