Alibaba Group’s research institute DAMO Academy has developed an AI-powered tool that can screen for early signs of pancreatic cancer, one of the most fatal cancers worldwide.
The tool, dubbed PANDA by researchers, targets pancreatic ductal adenocarcinoma, the most common type of pancreatic cancer responsible for nearly half a million deaths globally every year, according to estimates.
“AI plus non-contrast CT technology hold the promise to be an effective and cost-efficient tool to achieve detection of pancreatic cancer in the early stages and make large-scale pancreatic cancer screening possible to prevent loss of lives,” said Le Lu, Head of Damo Academy’s medical AI team and a Fellow at the Institute of Electrical and Electronics Engineers.
PANDA is 34.1% more sensitive than radiologists to picking up abnormalities when it comes to screening scans, according to a study published last month by DAMO Academy in collaboration with over 10 medical institutions.
The algorithm was validated in a clinical setting during the screening of over 20,000 patients. While running the tests, PANDA detected pathological changes as a result of pancreatic cancer in 31 patients that doctors had missed.
The deep learning-based model is able to detect cancerous masses in the pancreas by examining non-contrast CT scans, a more efficient form of medical imaging used worldwide with a lower dose of radiation than contrast CT scans.
As well as being more sensitive than technicians, PANDA is also highly specific, meaning it produces very few false positive results. It has one false-positive in every 1,000 tests, outperforming radiologists by 6.3%.
Until now, early-stage detection of pancreatic cancer has been challenging, as symptoms often do not present until cancerous tumors in the organ are already large and spreading to other parts of the body.
But it’s potentially curable if caught early, with up to 10% of patients who receive an early diagnosis making a full recovery after treatment, according to research from Johns Hopkins University.
DAMO Academy’s screening tool can make this ideal scenario a reality by potentially integrating it into routine medical checkups or during visits to emergency departments. PANDA may one day even be able to detect other types of cancer.
“It is possible to envision a future in which AI is used to combine information from routine imaging with information on clinical history,” German medical researchers Jörg Kleeff and Ulrich Ronellenfitsch commented in an article on Nature Medicine.