I-TUM-CNN:Enhanced Intracranial Tumour Detection in MRI Images through CNN-based Analysis
Keywords:
Intracranial tumour, Brain lesion, Convolutional neural network , Magnetic reasoning imagingAbstract
INTRODUCTION: A brain lesion is a damaged piece of tissue, where tumours can be considered as a subset of lesions. Accurate diagnosis and classification of brain lesions play a vital role in understanding their progression.
OBJECTIVES: Magnetic Resonance Imaging (MRI) has emerged as a valuable medical imaging technique for identifying intracranial tumours. However, the manual interpretation of MRI images is time-consuming and requires expertise. In recent years, advancements in computer-assisted diagnosis (CAD), particularly machine learning and deep learning, have provided opportunities for radiologists to consistently identify intracranial tumours.
METHODS: Traditional machine learning methods often rely on handcrafted features for classification. In contrast, deep learning approaches offer the advantage of automating feature extraction without the need for manual intervention, resulting in accurate classification results. This study presents a model that utilises machine learning techniques to accurately detect brain lesions, including tumours, in magnetic resonance imaging. The algorithm employed a Convolutional Neural Network (CNN) for feature extraction and segmentation.
RESULTS: The dataset used for training and evaluation was obtained from an online source. The results demonstrate the promising performance of the proposed model, achieving an impressive accuracy of 97.79%.
CONCLUSION: Using deep learning techniques in conjunction with MRI analysis holds great potential for improving the efficiency and accuracy of lesion detection in clinical settings.