A Comprehensive Review of Multi-classification of Brain Tumour MRI Images using Deep Dynamic Parallel Convolutional Neural Network with Fully Termite Alate Optimization Algorithm

Abstract

Brain tumour classification using magnetic resonance imaging (MRI) plays a vital role in early diagnosis, treatment planning, and patient survival. Traditional diagnostic approaches rely heavily on manual interpretation by radiologists, which can be time-consuming and prone to variability. Recent advancements in deep learning, particularly convolutional neural networks (CNNs), have significantly improved automated brain tutor detection and classification. CNN-based models enable automatic feature extraction and classification of tumours into multiple classes such as glioma, meningioma, pituitary tumour, and normal tissues. Studies indicate that deep learning approaches achieve high accuracy exceeding 95% in multi-class MRI classification tasks. Furthermore, hybrid architectures such as dynamic parallel CNNs enhance feature representation by processing multiple convolutional pathways simultaneously. Optimization algorithms play a critical role in improving model performance, where bio-inspired techniques such as Termite Alate Optimization (TAO) provide efficient global search capabilities and optimal hyperparameter tuning. These algorithms help avoid local minima and improve convergence speed. This paper presents a comprehensive review of deep dynamic parallel CNN frameworks integrated with Termite Alate Optimization for multi-class brain tutor classification. The study explores recent advancements, challenges, and future research directions in AI-driven neuroimaging, highlighting the importance of hybrid deep learning and optimization techniques for improving diagnostic accuracy and clinical applicability.