New tool finds deadliest cancer years before tumors seen on scan
Key Points:
- Researchers at the Mayo Clinic have developed an AI model capable of detecting pancreatic cancer abnormalities on CT scans up to three years before diagnosis, potentially improving early detection of this aggressive cancer.
- The AI outperformed radiologists by identifying early signs of pancreatic cancer three times more effectively, including abnormal pancreatic cells that evade immune defenses, which have been difficult for doctors to detect.
- Pancreatic cancer has a low five-year survival rate of 13% and causes over 52,700 deaths annually; early detection is crucial as symptoms are often vague and mistaken for less serious conditions.
- The AI tool is currently in clinical trials and aims to assist individuals with a family history of pancreatic cancer who do not yet show symptoms, offering hope for earlier intervention.
- Despite advancements in AI for medical advice, experts caution that AI cannot replace the judgment and ethical accountability of healthcare professionals, as some studies show patients may prefer AI advice even when it is inaccurate.