Y-net: a reducing gaussian noise convolutional neural network for MRI brain tumor classification with NADE concatenation. (13th July 2021)

• Record Type: Journal Article
• Authors: Hashemzehi, Raheleh; Seyyed Mahdavi, Seyyed Javad; Kheirabadi, Maryam; Kamel, Seyed Reza
• Is Part Of: Biomedical physics & engineering express. Volume 7:Number 5(2021)
• Publication Date: 2021-07-13
• Subjects: NADE -- convolutional neural network -- Y-net -- U-net -- brain tumor classification; Medical physics -- Periodicals; Biophysics -- Periodicals; Biomedical engineering -- Periodicals; Medical sciences -- Periodicals; 610.153
• DOI: 10.1088/2057-1976/ac107b
• Languages: English
• ISSNs: 2057-1976

Abstract

Abstract: Brain tumors are among the most serious cancers that can have a negative impact on a person's quality of life. The magnetic resonance imaging (MRI) analysis detects abnormal cell growth in the skull. Recently, machine learning models such as artificial neural networks have been used to detect brain tumors more quickly. To classify brain tumors, this research introduces the Y-net, a new convolutional neural network (CNN) based on the convolutional U-net architecture. We apply a NADE concatenation method in pre-processing the MR images for enhanced Y-net performance. We put our approach to the test using two MRI datasets of brain tumors. The first dataset contains three different types of brain tumors, while the second dataset includes a separate category for healthy brains. We show that our model is resistant to white noise and can obtain excellent classification accuracy with a limited number of medical images.