A novel approach for early gastric cancer detection using a hybrid of alexnet, extreme learning machine, and adjusted gorilla troops optimization

Abstract

Gastric cancer is a prevalent and highly lethal form of cancer that affects individuals worldwide. The early detection of gastric cancer plays a vital role in ensuring successful treatment and improved patient outcomes. However, identifying this malignancy at an early stage can be challenging, and conventional diagnostic techniques often exhibit limited accuracy. To address this issue, a groundbreaking study proposes a novel methodology that combines the utilization of the Alexandria Net (Alexnet), Extreme Learning Machine (ELM), and a modified metaheuristic algorithm. The integration of these techniques aims to enhance the accuracy and efficiency of gastric cancer detection. Additionally, a Modified version of the Dynamic Gorilla Troops Optimizer (DGTO) is incorporated as a dynamic metaheuristic algorithm. This algorithm optimizes the arrangement and selection of features within the model, significantly contributing to improved performance. To evaluate the effectiveness of the proposed methodology, the Alexnet/ELM-DGTO model is applied to a practical dataset known as GasHisSDB. Moreover, this approach outperforms several state-of-the-art methods in terms of precision, accuracy, recall, specificity, F1-score, and Area Under the Curve (AUC). This research paper makes a significant contribution to the field of early gastric cancer detection by introducing a novel methodology that overcomes the limitations of existing approaches. It offers a unique solution to the optimization problem of selecting the most suitable features for machine learning models. The effectiveness of the proposed methodology is further validated through its successful application on a practical dataset, highlighting its superiority over alternative methods. The results underscore the potential impact of utilizing the Alexnet/ELM-DGTO combination in the field of gastric cancer detection.