What is it about?
Brain Hemorrhage is bleeding inside the skull when thin brain vessels are teared. We have created a deep learning model and also made comparisons with other established deep learning models to automatically detect the hemorrhage from CT scan images. We worked on CT scan images as they are still very widely used and cheap in all countries.
Featured Image
Photo by National Cancer Institute on Unsplash
Why is it important?
Our work prevents misdiagnosis as it has an accuracy of 96%. It is completely automatic and can detect hemorrhage within seconds. The model works on CT scan images as it widely available and is affordable for all people.
Perspectives
Working on this topic was very interesting and it inspired me and my co-authors as we were able to contribute to health research. Since CT scan images are still manually read by humans, it is time consuming and can, at times, lead to misreading the images, hence in wrong diagnosis. Our model can detect images at accuracy of 96% and is very fast compared to manual reading. As it automatically read, we can get instant results.
Tasin Al Nahian Khan
BRAC University
Read the Original
This page is a summary of: Intracranial Hemorrhage Detection on CT Scan Images using Transfer Learning Approach of Convolutional Neural Network, March 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3542954.3542980.
You can read the full text:
Resources
Contributors
The following have contributed to this page
