Quantum Bat Algorithm Using Ensemble Bayesian Neural Networks for Cad-Based Lung Cancer Prediction

Abstract

Lung malignancies have raised global death rates, necessitating early therapies. This uses Computer Tomography (CT) pictures to identify lung cancer. The first stage is to automatically split Regions of Interest (ROI) when cancerous spots are discovered in order to increase diagnosis Accuracy (AC). When maximizing feature extractions, the Quantum inspired Bat Optimization Algorithm (QBOA) considers only relevant attributes. Ensemble Bayesian Neural Networks (EBNNs), which are used in medical applications to categorize data, may make good predictions while providing imprecise estimations. The LIDC-IDRI dataset was used in this work to verify the suggested approach, which is highly accurate in detecting and categorizing stages of lung cancer. Lung malignancies may be successfully identified using segmentations, optimum feature extractions, and ensemble-based classification, allowing for early intervention and improved patient care. Furthermore, the outcomes indicate how quickly and effectively the framework can detect and diagnose lung cancer. This unique technology offers a trustworthy and effective option for diagnosing early lung cancer, opening the path for improved clinical decision-making and patient outcomes.