Brain Tumor Classification Using Deep Learning Technique - A Comparison between Cropped, Uncropped, and Segmented Lesion Images with Different Sizes

Abstract

Deep Learning is the newest and the current trend of the machine learning field that paid a lot of the researchers' attention in the recent few years. As a proven powerful machine learning tool, deep learning was widely used in several applications for solving various complex problems that require extremely high accuracy and sensitivity, particularly in the medical field. In general, the brain tumor is one of the most common and aggressive malignant tumor diseases which is leading to a very short expected life if it is diagnosed at a higher grade. Based on that, brain tumor grading is a very critical step after detecting the tumor in order to achieve an effective treating plan. In this paper, we used Convolutional Neural Network (CNN) which is one of the most widely used deep learning architectures for classifying a dataset of 3064 T1 weighted contrast-enhanced brain MR images for grading (classifying) the brain tumors into three classes (Glioma, Meningioma, and Pituitary Tumor). The proposed CNN classifier is a powerful tool and its overall performance with an accuracy of 98.93% and sensitivity of 98.18% for the cropped lesions, while the results for the uncropped lesions are 99% accuracy and 98.52% sensitivity and the results for segmented lesion images are 97.62% for accuracy and 97.40% sensitivity.

Recognition of Handwritten Arabic and Hindi Numerals Using Convolutional Neural Networks

“…It is commonly used in multivariant classifications purposes. The fully connected layer has nodes with the same number of output classes [38]. In the paper, we utilized the properties in MATLAB ® 2019 [39] to build our own CNN which consists of 18 layers and employs it in discriminating the numerals.…”

The Application of Deep Convolutional Neural Networks to Brain Cancer Images: A Survey

“…Alqudah et al [ 36 ] put forth a new DCNN architecture with just 18 layers for classifying (grading) MR images into three classes of tumors including Meningioma, Glioma, and Pituitary. They used a public dataset containing 3064 brain MR images (T1 weighted contrast-enhanced).…”

A Systematic Approach for MRI Brain Tumor Localization and Segmentation Using Deep Learning and Active Contouring

“…Due to the easy accessibility and the ready availability, the Figshare MRI brain tumour dataset also has been used in many brain tumor classification and segmentation related research [15][16][17][18]. e dataset, which was initiated in 2015 and last updated in 2017 [13,16], carries an average classification accuracy in the range of 90-95% [14,16,19,20]. e authors in [16] achieved a classification average of 95% accuracy by using a modified CNN architecture while the authors in [15] achieved around 96% accuracy with an automatic content-based image retrieval (CBIR) system.…”