A newly developed AI model from researchers at the Mayo Clinic and the University of Texas MD Anderson Cancer Center indicates that AI can now detect pancreatic cancer several years prior to diagnosis.
Their innovative system, known as REDMOD (radiomics-based early detection model), was evaluated using CT scans from individuals who were subsequently diagnosed with pancreatic cancer.
In nearly 75% of the cases, REDMOD successfully identified the most prevalent type of pancreatic cancer approximately 16 months before diagnosis. This detection rate is nearly double that of specialists reviewing the scans without the aid of AI.
In certain scans, REDMOD recognised suspicious tissue patterns more than two years before diagnosis, and the research team believes it could potentially detect cancer up to three years in advance.
A radiologist and nuclear medicine specialist at the Mayo Clinic, Ajit Goenka has said, “The greatest barrier to saving lives from pancreatic cancer has been our inability to see the disease when it is still curable.”
“This AI can now identify the signature of cancer from a normal-appearing pancreas, and it can do so reliably over time and across diverse clinical settings”, the specialist added.
The researchers used 969 CT scans of the pancreas as training data for REDMOD, enabling it to learn how to identify the subtle indicators of cancer in its earliest stages.
According to the new study, there remains work to be done before this can be fully realised. The next step for the researchers is to test the AI on larger, more diverse populations and assess how easily it can be integrated into existing medical practices.
The researchers are optimistic about these preliminary findings and hope that with additional development and refinement, we may have an exceptionally valuable system for one of the most lethal cancers.
Read more: Scientists reveal major treatment that cuts bad cholesterol by 50%
