Researchers at the University of Gothenburg have developed an AI model, named Candycrunch, that can detect cancer through sugar analyses by mass spectrometry. The AI model is faster and more accurate in identifying abnormalities than the current semi-manual method. It has been trained using a database of over 500,000 examples and can determine the exact sugar structure in 90% of cases in just a few seconds per test. This automation can accelerate the discovery of glycan-based biomarkers for cancer diagnosis and prognosis, potentially revolutionizing the field of biological and clinical research.
Source link
AI model finds the cancer clues at lightning speed
