Researchers in New York have come up with a badass AI tool for predicting how long a patient diagnosed with cancer will survive. This machine learning tool is the real deal, folks. It works its magic specifically for breast, thyroid, and pancreatic cancers.
Now, here’s the kicker: this cancer survival calculator prototype goes beyond the typical cancer stage to make its prognosis. Lead study author Lauren Janczewski knows there are a ton of other factors that can affect a patient’s survival, and she wanted to take them into account. That’s why she and her team at Northwestern University McGaw Medical Center developed this kickass Cancer Survival Calculator to give patients a more personalized estimate of their cancer prognosis.
Using some sick AI called machine learning, they created this awesome tool and tested it using a nationwide cancer dataset. The goal was to figure out what characteristics have the biggest impact on survival for each specific cancer type. They gathered data from patients diagnosed between 2015 and 2017 with breast, thyroid, and pancreatic cancers. Then, they trained their machine learning algorithms to spot patterns between these characteristics and the patients’ survival at five years.
When it came time to test the calculator, they included data from a whopping number of patients – 259,485 with breast cancer, 76,624 with thyroid cancer, and 84,514 with pancreatic cancer. And guess what? The calculator was ridiculously accurate, predicting survival rates within just nine to ten months of the actual survival time.
But here’s the coolest part: this one-of-a-kind calculator takes into account specific tumor biomarkers and treatment variables that can affect a patient’s prognosis. That’s something a lot of other calculators miss the mark on.
Now, the next step is to get this bad boy into action. The researchers are working on a user interface that’ll make it easy for clinicians to use the Cancer Survival Calculator in their practice. They plan to test it out at selected cancer centers because they want to make sure it’s the real deal.
So, keep an eye out, my friends. This groundbreaking study was presented at the American College of Surgeons (ACS) Clinical Congress 2023, and I have a feeling it’s going to change the game when it comes to predicting cancer survival rates.