Intracranial hemorrhage detection in human brain using deep learning

A serious illness, Intracranial Hemorrhage (ICH) may result in severe impairment or death if not treated quickly. Different causes, ranging from trauma to vascular illness to congenital development, may result in this condition. ICH is further subdivided into epidural haemorrhage (EDH), subdural hae...

Full description

Saved in:
Bibliographic Details
Main Authors: Revathi, Ch. Bhanu, Kumar, J. M. S. V. Ravi, Sujatha, B.
Format: Conference Proceeding
Language:English
Subjects:
Algorithms
Bleeding
Hemorrhage
Illnesses
Image segmentation
Machine learning
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A serious illness, Intracranial Hemorrhage (ICH) may result in severe impairment or death if not treated quickly. Different causes, ranging from trauma to vascular illness to congenital development, may result in this condition. ICH is further subdivided into epidural haemorrhage (EDH), subdural haemorrhage (SDH), subarachnoid hemorrhage (SAH), cerebral parenchymal haemorrhage (CPH), and intraventricular haemorrhage (IVH) based on the site of bleeding (IVH). There are many kinds of bleeding, and the intensity of the bleeding and the treatments required differ. A new framework for automatic and accurate ICH identification is being developed as part of this research. It made use of CNN, which was utilized to extract relevant characteristics from picture slices and identify the kind of ICH present in that image slice using segmentation techniques. On the basis of our suggested algorithm's performance, we shall assess and compare it to CNN. To give visual proof of detection, a visualization method was also suggested that does not need the manual demarcation of bleeding regions for purposes of training. Detecting bleeding and identifying the right subtype of haemorrhage are our primary goals in this investigation.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0131258