IIALS: A Novel Methodology to Predict Lung Nodule Cancer on Earlier Stages by Using Integrated Image Assistive Learning Strategy

Lung cancer is one of the leading causes of cancer-related deaths globally, and early detection of malignant lung nodules is crucial for improving patient survival rates. This study presents a novel methodology, IIALS (Integrated Image Assistive Learning Strategy), which uses an ensemble of EfficientN et and Support Vector Machines (SVM) to enhance the prediction of lung nodule malignancy at early stages. The proposed model was tested on two widely used datasets, LUNA16 and LIDC-IDRI, and demonstrated superior performance compared to other state-of-the-art methods. The EfficientNet architecture was employed for robust feature extraction, while SVM was used for classification, ensuring a high level of precision. The model achieved an accuracy of 96.8% on the LUNA16 dataset and 95.4% on the LIDC-IDRI dataset, outperforming existing methods such as ResNet50 and Faster R-CNN. Key metrics such as sensitivity (97.2% on LUNA16), specificity (96.4% on LUNA16), and F1-Score (97.0% on LUNA16) further highlight the effectiveness of the proposed approach. The use of attention mechanisms, multi-scale feature extraction, and hybrid classification techniques contributed to these impressive results. This approach demonstrates the potential of advanced deep learning (DL) techniques in improving lung cancer detection, offering a reliable tool for clinical application.