A STUDY ON THE HYBRID APPROACH FOR LUNG CANCER DETECTION USING CONVOLUTIONAL NEURAL NETWORKS AND DECISION TREES

  • Unique Paper ID: 164201
  • Volume: 10
  • Issue: 12
  • PageNo: 786-801
  • Abstract:
  • Lung cancer detection is a critical aspect of early diagnosis and treatment, significantly impacting patient outcomes and survival rates. In recent years, the integration of advanced machine learning techniques such as Convolutional Neural Networks (CNNs) and decision trees has shown promising results in improving the accuracy and efficiency of lung cancer detection from medical imaging data. This review article provides an overview of recent research articles focusing on the application of CNNs and decision trees for lung cancer detection, highlighting key methodologies, findings, and future directions. Lung cancer remains one of the most prevalent and deadly forms of cancer worldwide, necessitating advanced diagnostic tools for early detection and effective treatment. In this study, we propose a hybrid approach utilizing Convolutional Neural Networks (CNNs) in conjunction with decision trees for accurate and efficient lung cancer detection from medical imaging data. The proposed methodology begins with the preprocessing of diverse lung image datasets, including computed tomography (CT) scans and X-rays, to standardize resolution and enhance image quality. Subsequently, a CNN architecture is employed to automatically extract relevant features from the preprocessed images. The CNN model comprises multiple convolutional layers followed by pooling layers to progressively learn hierarchical representations of lung nodules and surrounding tissues. Transfer learning techniques are also utilized to leverage pre-trained CNN models, thereby enhancing model generalization and performance on limited datasets. Following feature extraction, the learned representations are fed into decision tree classifiers for final diagnosis. Decision trees offer interpretability and transparency, enabling clinicians to understand the decision-making process underlying lung cancer detection. Ensemble learning techniques such as random forests are also explored to improve classification accuracy and robustness. To evaluate the effectiveness of the proposed hybrid approach, extensive experiments are conducted on benchmark lung cancer datasets, including publicly available repositories and clinical datasets. Performance metrics such as accur

Related Articles