Lung Tumor Segmentation Using Visual Geometry Group Networks In MRI Images

Lung Cancer Is One Of The Most Life-threatening Diseases Worldwide, And Early Diagnosis Significantly Increases The Chances Of Survival. Accurate Detection And Classification Of Lung Tumors Remain Challenging Due To Variations In Tumor Size, Shape, And Texture In MRI Images. This Study Proposes An Automated Lung Tumor Segmentation And Classification Framework Using Convolutional Neural Networks (CNN) And The Visual Geometry Group (VGG) Network Architecture. The Proposed System Utilizes MRI Images To Analyze Textural And Spatial Features Of Lung Tissues For Distinguishing Between Normal And Malignant Cases. A Multi-scale Feature Extraction Approach Is Incorporated To Improve Detection Performance And Enhance Classification Accuracy. The VGG Network Serves As The Base Model For Deep Feature Learning, While CNN Layers Refine Segmentation And Classification Tasks. The Developed Database Includes Multiple MRI Views To Ensure Robust Training And Validation. Experimental Results Demonstrate That The Proposed Model Achieves High Precision And Overall Classification Accuracy Of Up To 98%, As Evaluated Using Confusion Matrix Metrics. The System Reduces Manual Interpretation Errors And Provides An Efficient Computer-aided Diagnostic Tool For Early Lung Tumor Prediction.