DISEASE PREDICTION AND CLASSIFICATION OF LUNGS CANCER USING DEEP MACHINE LEARNING

Abstract

Lungs cancer is the leading cause of death. However, with about 2.5 million occurrences worldwide in 2022, colon cancer ranks as the third most prevalent disease. Globally, there would be 9.95 million cancer-related deaths and around 19.2 million new cases in 2024. Cancer poses a significant threat due to its aggressive nature, potential for widespread metastasis, and inherent he erogeneity, which often leads to resistance to chemotherapy. Lungs cancer ranks among the most prevalent of cancer worldwide, affecting individuals of all genders. Timely and accurate lungs cancer detection is critical for improving cancer patients’ treatment outcomes and survival rates. Screening examinations for lungs cancer detection, however, frequently fall short of detecting small polyps and cancers. To address these limitations, computer-aided techniques for lungs cancer detection proved as an invaluable resource for both health care practitioners and patients. This study implements an enhanced EfficientNetB1 deep learning model for accurate detection and classification using histopathological images. The proposed technique accurately classifies the histopathological images into three distinct classes: No cancer (benign), Adenocarcinomas, and squamous cell carcinomas. The performance of proposed technique is being evaluated by histopathological (LC25000) lungs dataset. The image resizing and augmentation are followed by loading a pre-trained model and applying transfer learning. The dataset is then split into training and validation sets, with fine-tuning and retraining performed on the training dataset. The proposed model performance is evaluated on the validation dataset, and the results of lungs cancer detection and classification achieved 99.8% accuracy.