A new AI tool developed by Amsterdam UMC could help general practitioners detect lung cancer up to four months earlier than current methods, offering a major improvement in survival rates and lowering treatment costs. Using data from over 500,000 patients, the AI algorithm examines both structured medical records and unstructured notes taken during routine GP visits. By finding subtle patterns, such as mild recurring symptoms or appointment trends, the tool can spot early signs of cancer before patients are usually referred for further testing.
Tested on real-world data from general practices across the Netherlands, the AI system was able to predict lung cancer diagnoses months earlier than traditional methods. Early-stage lung cancer is often much more treatable, and earlier detection can significantly increase the chances of survival. Unlike national screening programs, this AI tool can be used during a regular consultation without needing extra tests, and it appears to produce fewer false positives, making it practical and efficient for daily healthcare use.
Although the early results are very promising, researchers stress that more work is needed to further refine the tool and validate its effectiveness across different healthcare systems. The team also believes that the same technology could be adapted to detect other difficult-to-diagnose cancers, such as pancreatic or ovarian cancer. If successful, this AI tool could revolutionize how doctors detect cancers early, leading to better patient outcomes and a major shift in cancer diagnosis practices.
