AI Advancements in Cancer Screening
There’s an ongoing debate about whether AI is beneficial for humanity. However, recent developments in AI tools could significantly change cancer screening methods.
Pancreatic cancer is notably aggressive and notoriously hard to detect. It has a five-year survival rate of just 13% and is expected to result in over 52,700 deaths this year, as indicated by the American Cancer Society.
A new AI model created at the Mayo Clinic might help improve these alarming statistics by detecting abnormalities in CT scans as much as three years prior to a pancreatic cancer diagnosis.
The findings were published in the journal Gut. The study analyzed scans from patients screened for other illnesses who later received a diagnosis of pancreatic cancer.
In this study, both a radiologist and the AI model analyzed the scans for early signs of cancer. The results showed that the AI was three times more effective in spotting signs that are typically missed by the human eye.
Dr. Diane Simeone, director of the Moores Cancer Center at UC San Diego Health, pointed out that early symptoms like back pain and fatigue can often be mistaken for more benign issues. This ambiguity makes early detection incredibly challenging.
Pancreatic cancer originates from abnormal cell growth in a critical organ involved in digestion and blood sugar control.
Interestingly, the AI was also successful in identifying abnormal pancreatic cells, which have been difficult for scientists and physicians to detect in the past.
According to Dr. Ajit Goenka of the Mayo Clinic, “We knew this signal was there; we just needed a method to identify it.” The team is optimistic that this tool, currently in clinical trials, will assist individuals with a family history of pancreatic cancer who may not display symptoms.
While a long-term cure for pancreatic cancer has yet to be discovered, available treatments like chemotherapy, surgery, targeted therapy, and immunotherapy can help in managing the condition.
In related AI news, a study in the New England Journal of Medicine found that many people seem to prefer medical advice from AI bots over that of doctors or online platforms, even though the AI isn’t always accurate.
However, some experts advise caution, emphasizing that AI cannot replace the critical thinking, ethical considerations, and real-world experience that medical professionals possess. Andy Kurzig, CEO of AI-powered search engine Pearl.com, stated that keeping people informed is crucial, serving as a safety net that can save lives.
