Alibaba’s PANDA AI detects pancreatic cancer early using CT scans

Alibaba’s PANDA leverages advanced AI to identify early signs of pancreatic cancer from standard CT scans, helping clinicians spot aggressive tumours sooner while reducing reliance on high-radiation imaging techniques.

Alibaba has unveiled an artificial intelligence-based medical tool that could significantly improve early detection of pancreatic cancer, a disease often diagnosed too late for effective treatment. Known as PANDA, the system analyses non-contrast CT scans to identify pancreatic ductal adenocarcinoma (PDAC), the most common and aggressive form of the disease.

Introduced in November 2024, PANDA was developed to overcome limitations in conventional screening methods. Contrast-enhanced CT scans, while effective, expose patients to higher radiation and are unsuitable for large-scale screening. PANDA’s ability to work with standard, low-radiation scans offers a safer alternative for early identification.

Strong performance in real-world testing

According to data shared by researchers, PANDA has demonstrated strong results during large-scale clinical evaluations in China. In real-world testing involving tens of thousands of patients, the system achieved sensitivity levels of around 93 percent, while maintaining near-perfect specificity. This allowed it to detect small and otherwise overlooked tumours, sometimes before patients developed noticeable symptoms.

The AI has also shown the ability to distinguish pancreatic cancer from conditions with similar imaging features, such as cysts and pancreatitis, reducing the risk of misdiagnosis. Doctors using the system reported that it flagged several hidden cancer cases during routine scans, enabling earlier clinical intervention.

Following screenings conducted on more than 70,000 individuals in China, PANDA is now being prepared for regulatory review in the United States. If approved, the technology could broaden access to early pancreatic cancer detection, improving survival prospects and reducing diagnostic uncertainty for patients worldwide.