Early forecast
Artificial intelligence detects the first signs of pancreatic cancer
An artificial intelligence (AI) tool developed by Cedars-Sinai researchers has been able to predict who will develop pancreatic cancer. AI analyzed CT scans years before patients were diagnosed. New technology will help diagnose pancreatic cancer earlier, reports Cedars-Sinai. The results of the work are published in the journal Cancer Biomarkers (Qureshi et al., Predicting pancreatic ductal adenocarcinoma using artificial intelligence analysis of pre-diagnostic computed tomography images).
"This AI tool was able to record and quantify very subtle early signs of pancreatic duct adenocarcinoma on computed tomography years before the onset of the disease. These are signs that the human eye will never be able to distinguish," comments senior study author Debiao Li, director of the Biomedical Imaging Research Institute, professor of Biomedical sciences and imaging at Cedars—Sinai.
Pancreatic duct adenocarcinoma is not only the most common type of pancreatic cancer, but also the most dangerous. People with this disease may experience symptoms such as general abdominal pain or unexplained weight loss, but there are no unique early signs that accurately identify pancreatic cancer.
In this study, researchers examined electronic medical records to identify people who had been diagnosed with cancer over the past 15 years and who had undergone CT scans six months to three years before diagnosis. These CT images were considered normal: no signs of tumor formation were detected.
The authors of the work taught the AI tool to analyze these prediagnostic CT images of people with pancreatic cancer and compare them with CT images of 36 people who did not develop cancer. The researchers reported that the model identified people who will eventually be diagnosed with pancreatic cancer, and those who will not develop cancer, with an accuracy of 86%. An artificial intelligence model has identified differences on the surface of the pancreas in cancer patients and in healthy people.
Portal "Eternal youth" http://vechnayamolodost.ru