Enhanced Brain Tumour Detection Using Adaptive Preprocessing and Hybrid Deep Learning Techniques

Abstract

Magnetic Resonance Imaging (MRI) is a crucial diagnostic tool for neurological and brain cancers, which are major global health issues. However, hand-analysed MRI scans are inconsistent and time-consuming. This paper describes a sophisticated strategy for employing Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) to detect brain cancers. By combining hybrid Deep Learning (DL) algorithms with superior preprocessing, the framework improves performance. Initially, Adaptive Multi-scale Gaussian Filtering (AMGF) is applied to minimize noise and improve image quality. Following the identification of the most relevant features using Recursive Feature Elimination with Cross-Validation (RFECV), Lévy Flight Particle Swarm Optimization (LFPSO) is used to optimize feature extraction. To perform classification, a hybrid DL model combines Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks, identifying both temporal and spatial patterns accurately. This technique provides significant increases in detection accuracy and processing performance. indicating its potential for clinical use and better patient outcomes. The results shown that the proposed method has the best results around the comparison methods.