Detection and Classification of Brain Hemorrhage Based on Hounsfield Values and Convolution Neural Network Technique

In recent years, brain hemorrhage tends to increase rapidly and is one of the most dangerous for our life. Thus, the automatic detection and classification of brain hemorrhage are essential for doctors to treat. In this paper, we present a new approach using convolution neural network technique and...

Full description

Saved in:
Bibliographic Details
Published in2019 IEEE-RIVF International Conference on Computing and Communication Technologies (RIVF) pp. 1 - 7
Main Authors PHAN, Anh-Cang, NGUYEN, Thi-My-Nga, PHAN, Thuong-Cang
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.03.2019
Subjects
Biological neural networks
Brain hemorrhage
Brain modeling
Convolution
convolution neural network
CT/MRI images
Deep learning
Feature extraction
Hemorrhaging
Hounsfield Unit
Image segmentation
Online AccessGet full text

Cover

More Information
Summary:In recent years, brain hemorrhage tends to increase rapidly and is one of the most dangerous for our life. Thus, the automatic detection and classification of brain hemorrhage are essential for doctors to treat. In this paper, we present a new approach using convolution neural network technique and Hounsfield Unit values for analyzing brain hemorrhage from CT/MRI images. The proposed method includes two stages: classification of brain hemorrhage using convolution neural network technique; detection of brain hemorrhage areas and determination of brain hemorrhage time based on Hounsfield Unit values. Our method is effective for doctors in recognizing brain hemorrhage, diagnosing the location, time and severity level of brain hemorrhage for prompt patient treatment.
DOI:10.1109/RIVF.2019.8713733