Recent studies have shown that patients with early-detected pancreatic ductal adenocarcinoma (PDAC) have a median overall survival of 9.8 years, compared to 1.5 years for late diagnoses. Known as the “king of cancers,” pancreatic cancer has gained notoriety in recent years due to the tragic losses of high-profile figures such as Apple’s Steve Jobs and Wu Zunyou, chief scientist at the Chinese Centre for Disease Control and Prevention.
A game-changing achievement in early-stage screening for this deadly cancer has been made possible through the development of an artificial intelligence tool Chinese scientists. This pioneering screening model specifically targets PDAC, the predominant subtype responsible for over 95 percent of pancreatic cancer cases. With an impressive specificity of 99.9 percent and sensitivity surpassing the average radiologist’s performance 34.1 percent, the model shows great promise but requires further regulatory approval for practical implementation.
Lead author Cao Kai and Lu Le, the leader of DAMO Academy’s medical team, conceived the idea of using AI for early cancer screening to address the inadequate existing early screening tools. They initiated a research project with over 10 medical institutions to develop a technology combining non-contrast CAT scans with AI for large-scale pancreatic cancer screening, resulting in the creation of an algorithm named “Pancreatic Cancer Detection with Artificial Intelligence,” abbreviated as PANDA.
Benefiting from an extensive dataset, rigorous data processing, and a pioneering training strategy, PANDA emerged as a highly perceptive AI imaging expert. The study revealed PANDA’s efficient detection of lesions in a multi-center validation cohort, surpassing a radiologist’s average performance 6.3% in specificity and 34.1% in sensitivity for pancreatic lesions. PANDA achieved 92.9% sensitivity and 99.9% specificity in a large-scale validation, demonstrating its effectiveness for detecting and diagnosing pancreatic lesions and showcasing its potential for large-scale screening and early detection of PDAC.