LUNG CANCER PREDICTION WITH CNN

Computer Science and Engineering

Lung cancer remains a leading cause of cancer-related mortality worldwide, underscoring the critical need for accurate and timely diagnosis. In recent years, deep learning techniques, particularly Convolutional Neural Networks (CNNs), have emerged as powerful tools for medical image analysis and disease prediction. This paper presents a comprehensive analysis of a CNN-based approach for predicting lung cancer using medical imaging data. Our methodology begins with the preprocessing of lung scan images, where we standardize their dimensions and introduce variations through Gaussian blurring. To address class imbalance inherent in medical datasets, we apply Synthetic Minority Over-sampling Technique (SMOTE) to ensure balanced representation across benign, malignant, and normal cases.The custom CNN architecture, constructed using Keras, is optimized for learning features relevant to lung cancer prediction. Trained from scratch, our model demonstrates its efficacy in discriminating between different types of lung conditions, achieving promising accuracy on a validation set. Our findings suggest that our approach holds significant promise for early lung cancer prediction, potentially revolutionizing diagnostic practices and improving patient outcomes. This research contributes to the ongoing efforts in combating lung cancer and underscores the importance of leveraging deep learning and data augmentation techniques in medical imaging analysis.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.