Feature Dimensionality Reduction for Lung Tumor Classification Using Transformer Deep Learning and YOLO Model

Healthcare ranks among the most important sectors of any economy because of the impact it has on the whole country.The discovery is going to help the healthcare industry save money by lowering the expenses of lung cancer diagnostics.This research aims to find appropriate feature transformation methodologies using dimensionality reduction techniques and an acceptable regression model that can robustly execute this task.It uses the lung cancer dataset for early carcinoma diagnosis.In order to decrease diagnostic expenses and improve patient outcomes, early detection is essential.To accurately classify lung tumors utilizing demographic, clinical, and imaging data, this research offers a new deep learning framework called Compact Feature Set with Dual Ranking integrated with Transformer-based YOLOv5 (CFS-DR-TYOLOv5).For the purpose of learning long-range correlations within patient health data, the model uses a transformer-based architecture and incorporates dimensionality reduction techniques to extract compact and meaningful features from large-scale lung cancer datasets.Incorporating YOLOv5 allows for the accurate and quick detection of lung nodules in chest MRI images.The suggested model reduces computational complexity while increasing diagnostic accuracy through the combination of structured data and visual analysis.Based on the experimental results, CFS-DR-TYOLOv5 achieves better accuracy and precision in classification tasks compared to standard models for early-stage tumor identification.In order to improve clinical decisionmaking and early lung cancer screening, this integrated approach provides a dependable and scalable solution.