Artificial intelligence

in Early Cancer Detection and Risk Classification

A Cross-Disciplinary National Research Project to Improve HPB Cancer Treatment

Artificial intelligence and deep learning have shown promising results in the early detection of tumors from radiological imaging and even in disease classification. AI refers to a machine's (or algorithm's) ability to perform tasks traditionally associated with human intelligence, such as learning. By using AI methods, we can develop learning algorithms that give a system designed for a specific application the ability to make observations, process data, and achieve specific goals.

In our research, CT-scan material from the Finnish Hepato-Pancreato-Biliary Cancers Cohort (FinHPB) is used to develop AI algorithms for the early detection and risk classification of pancreatic tumors. The FinHPB cohort includes all patients with liver, pancreatic, and biliary tract cancers who underwent surgery in Finland between 2000 and 2019, totaling 6,930 patients.