Researchers at Cedars-Sinai have created a cutting-edge Artificial Intelligence (AI) tool that can accurately diagnose early signs of pancreatic cancer.
The novel AI technology effectively identifies signs of pancreatic cancer by analysing a patient’s CT scan images. The tool has been demonstrated to predict who will develop pancreatic cancer years before they were diagnosed with the disease, potentially providing a vital tool for detecting one of the most difficult-to-treat cancers.
Debiao Li, PhD, director of the Biomedical Imaging Research Institute, professor of Biomedical Sciences and Imaging at Cedars-Sinai, and senior and corresponding author of the study, commented: “This AI tool was able to capture and quantify very subtle, early signs of pancreatic ductal adenocarcinoma in CT scans years before the occurrence of the disease. These are signs that the human eye would never be able to discern.”
The findings of the research are published in the journal Cancer Biomarkers.
What is pancreatic ductal adenocarcinoma?
Pancreatic ductal adenocarcinoma is the most deadly and common form of pancreatic cancer, with less than 10% of people with the disease living longer than five years following a diagnosis or starting treatment. However, recent studies illuminated that detecting early signs of pancreatic cancer can elevate survival rates by 50%, although there is currently no easy method to achieve this.
Individuals with pancreatic ductal adenocarcinoma can experience abdominal pain or unexplained weight loss; however, these symptoms are often overlooked as they could infer a range of common health conditions, not necessarily signs of pancreatic cancer.
Stephen J. Pandol, MD, director of Basic and Translational Pancreas Research and program director of the Gastroenterology Fellowship Program at Cedars-Sinai, and another author of the study, commented: “There are no unique symptoms that can provide an early diagnosis for pancreatic ductal adenocarcinoma. This AI tool may eventually be used to detect early disease in people undergoing CT scans for abdominal pain or other issues.”
Locating early signs of pancreatic cancer
To design their AI tool, the researchers examined electronic health records to identify people diagnosed with pancreatic cancer within the last 15 years who had undergone CT scans between six months and three years before their diagnosis. From this data, the researchers found 36 patients who met the criteria, most of whom had CT scans due to abdominal pain.
Subsequently, the team trained the AI tool to analyse these pre-diagnostic CT images from pancreatic cancer patients and compare them with CT images from 36 people who did not develop the disease. The results showed that the AI technology had an accuracy of 86% in identifying early signs of pancreatic cancer and also those who would not develop the disease.
Moreover, the innovative system determined variations on the surface of the pancreas between people with cancer and healthy controls. These textural differences potentially occur due to molecular changes during the development of pancreatic cancer.
Touseef Ahmad Qureshi, PhD, a scientist at Cedars-Sinai and the first author of the study, concluded: “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.”
The researchers are now working on collating data from thousands of patients at US healthcare sites to further investigate the AI tool’s prediction capability.
