Year: 2025
Type: article
Abstract: Brain tumours constitute a substantial health issue owing to their intricate characteristics and elevated death rate. Timely and precise diagnosis is crucial for optimal treatment planning. Magnetic Resonance Imaging (MRI) is extensively used in medical imaging because of its superior contrast resolution, especially for soft tissues such as the brain. Nonetheless, the manual interpretation of MRI images is labour-intensive and susceptible to human mistake. This paper presents an improved method for brain cancer identification via Convolutional Neural Networks (CNNs), a category of deep learning models known for their exceptional efficacy in image classification and pattern recognition tasks. The proposed technique automates tumour identification and classification by learning complex characteristics directly from MRI images, therefore obviating the need for manually constructed features. Experimental findings indicate enhanced accuracy, sensitivity, and specificity relative to conventional machine learning methods, underscoring the promise of CNNs in aiding radiologists and enhancing therapeutic outcomes.
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7217970767
