Enhanced Superpixel Guided ResNet Framework with Optimized Deep Weighted Averaging Based Feature Fusion for Lung Cancer Detection in Histopathological Images

Lung cancer is a major health issue and a leading cause of cancer-related mortalities globally. Early diagnosis is essential for improving survival rates, with biopsy as the gold standard for tissue analysis. While digital histopathology enhances image quality and precision, manual analysis is time-consuming for pathologists, creating a need for automated classification methods. This research starts with image preprocessing using an adaptive fuzzy filter and segmentation via a Modified Simple Linear Iterative Clustering (SLIC) algorithm. The Segmented images are input to the Deep Learning architectures like ResNet-50 (RN-50), ResNet-101 (RN-101), and ResNet-152 (RN-152). Features extracted from these ResNet variants are fused using a Deep Weighted Averaging- Based Feature Fusion (DWAFF) technique, resulting in fused features termed ResNet-X (RN-X). To further refine these features, Particle Swarm Optimization (PSO) and Red Deer Optimization (RDO) techniques are employed within the Selective Feature Pooling layer. The optimized features are then passed to a Classification Layer that implements classifiers including Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor (KNN), SoftMax Discriminant Classifier (SDC), Bayesian Linear Discriminant Analysis Classifier (BLDC), and Multilayer Perceptron (MLP). Performance is assessed using K-fold cross-validation with K values of 2, 4, 5, 8, 10, and the results are compared using standard performance metrics. RN-X features obtained from the proposed DWAFF technique, combined with the MLP classifier, achieved a peak accuracy of 98.68% when using segmentation and RDO in the feature selection layer with K=10.