A team at Johns Hopkins University has developed an AI diagnostic tool that can detect signs of pancreatic cancer up to two years before traditional imaging methods would identify the disease. The system analyzes routine blood test results and medical history data to flag high-risk patients for further screening.
In clinical trials involving 25,000 patients, the AI system achieved a sensitivity of 92 percent and a specificity of 95 percent, dramatically outperforming current screening approaches. Early detection is critical for pancreatic cancer, which has a five-year survival rate of just 12 percent when caught late.
The FDA has granted the tool breakthrough device designation, fast-tracking it for regulatory approval. Researchers estimate it could save thousands of lives annually by catching cancers at treatable stages.
