Sudhakar Jha
... Read More
Written By: Sudhakar Jha | Updated : August 27, 2018 3:38 PM IST
Fighting the menace that's cancer has been a challenge for the mankind and any new discovery in technology seems as a potential solution to the problem. A new one has arrived, thanks to a team of computer engineers who have created an artificial intelligence system to help fight cancer.
Researchers from the University of Central Florida's Center for Research in Computer vision have created an AI that has the 'vision' to spot even the smallest of tumours in CT scans. To give you a perspective, radiologists have a success rate of 65 per cent in recognising smaller tumours on a scan, but this AI system has a 95 per cent accuracy rate, which will give the radiologists a new set of 'eyes' for scanning cancer tumours.
The AI systems needed to learn how the tumours look like and the different varieties of them in different parts of the body, and that's why the scientists used the brain as a model to create this particular AI system explained Rodney LaLonde, a doctoral candidate. LaLonde is also a computer science engineer and serves as the captain of UCF's hockey team. LaLonde was assisted by Engineering Assistant Professor Ulas Bagci.
"You know how connections between neurons in the brain strengthen during development and learn? We used that blueprint, if you will, to help our system understand how to look for patterns in the CT scans and teach itself how to find these tiny tumours," added LaLonde.
For developing the system, the team of researchers teamed up with the National Institute of Health and the Mayo Clinic and scanned thousands of CT scans into their AI software. The CT scans then taught the AI on how to look for tumour size and shape, among other trademarks.
The research team is hopeful that the new AI system will be installed in the hospitals soon and that will help increase the survival rates of the patients.
Image source: Shutterstock