Lungs cancer is one of the most common cancer among men and second most in women. The symptoms of lung cancer are coughing in blood, pain in chest and shortness in breath. In U.S lung cancer is responsible for more than 29 percent of cancer deaths.
For the prevention of prevention of this cancer and in order to detect it at its early stage, Researches at the University of Central Florida has created an AI system. This AI system is capable to detect small lung cancer tumors in computed tomography scans which are often very difficult to identify for radiologists.
“Detection of tiny/small objects has remained a very challenging task in computer vision, which so far has only been solved using computationally expensive multi-stage frameworks,” the researchers wrote.
“I believe this will have a very big impact,” Ulas Bagci, assistant professor of engineering and imaging scientist at the University of Central Florida, said in a UCF statement. “Lung cancer is the number one cancer killer in the United States and if detected in late stages, the survival rate is only 17 percent. By finding ways to help identify earlier, I think we can help increase survival rates.”
This AI system has given 95 percent accurate result in spotting lung cancer tumors in CT scan. If we compare it to human then the accuracy rate of a human eye is only 65 percent.
Researches at the University of Central Florida has trained this AI system over 1,000 CT scans that were publicly available from the Lung Nodule Analysis at National Institutes of Health. Each CT scan taught the AI what to look for in tumor size and shape, among other trademarks, and the system learned how to differentiate between cancerous and benign tumors.
Researches have created a deep learning algorithm name S4ND which help this AI system to detect small lung cancer tumors. S4ND doesn’t require any post-processing or user guidance to refine detection results. S4ND uses a single feed-forward pass of a single network for detection.
In the initial experiment to test the efficacy of S4ND, researchers compared this algorithm with the current leading method for lung nodule detection—3D convolutional neural network (3D CNN). The working approach of this algorithm to detect tiny specks of lung cancer in CT scans is similar to the algorithms of facial-recognition found on the iPhone X.
“We used the brain as a model to create our system,” which scans thousands of faces looking for a particular pattern to find its match, said Rodney LaLonde, a doctoral candidate at the UCF center.
LaLonde was assisted by Ulas Bagci, an assistant professor in Engineering, who leads the group of researchers in the center that focuses on AI with potential medical applications.
Researches said that to make this AI work more efficient with patience they will soon move the research project into a hospital setting. After that, the technology could be one or two years away from the marketplace, Bagci said.
The UCF team will present their study at the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI ), which takes place in Spain in September 2018.
Read also, you may like it :