Researchers from Cedars-Sinai Medical Centre in Los Angeles worked with an AI software that showed it could go a long way in early detection of pancreatic cancer.
Findings reveal that the technology had an 86 per cent accuracy in predicting if a patient would develop the disease. It studied over 108 CT scans from patients over the past 15 years who had been diagnosed with pancreatic cancer.
The patients CT scans were done six months to three years before diagnoses and it was the job of the AI software to see if there were any differences. Co-author of the study Dr Debiao Li explained how the software was able to make a diagnosis.
“We know in advance some image features are predictive of pancreatic cancer, “he said.
“However, they are very subtle and human eyes couldn’t readily discern them, so we use computers to find them.
“We select some of these features using data analysis tools to form the prediction model.
“Once the model is developed, we know what features are included in the prediction model and computers will find these features in new images to make a prediction of cancer or assess cancer risk.
“In fact, the system looks at 4,000 different features when it processes each scan.”
First author of the study Touseef Ahmad Qureshi said that early detection is the goal of this research.
“Our hope is this tool could catch the cancer early enough to make it possible for more people to have their tumour completely removed through surgery,” he said.
The hope for this technology is to predict pancreatic ductal adenocarcinoma (PDAC) which according to the study is one of the deadliest and most common forms of pancreatic cancer. The survival rate is low and symptoms such as a sore stomach, jaundice and weight loss are often over looked with most people receiving a diagnosis when it’s too late.
