Using AI To Decrease Spinal Fracture Risk After Cancer Diagnosis
Category: Spine | Author: Stefano Sinicropi
Technological advances are making it easier for medical experts to make significant advances in the field of patient care, and new artificial intelligence tools may soon be able to help doctors predict spinal fracture risk in patients battling cancer.
The advancement is important as more than 1.6 million Americans are diagnosed with cancer every year, and in roughly 10 percent of those cases, patients experience a spinal metastasis, where the disease spreads from another area into the spinal column. These new spinal tumors can lead to spinal instability or greatly increase spinal fracture risk, and spinal fractures increase a patient’s risk of death by about 15 percent. However, artificial intelligence may soon be able to better predict spinal fracture risk as a result of spinal tumor growth so that we can help protect patients from a problematic fracture.
The new study used an AI-assisted framework called ReconGan to create a virtual reconstruction of a patient’s vertebral column. Unlike 3D printing, where a virtual model is turned into a physical object, this “digital twin” uses real-life data to create a computer model without a physical creation. Researchers can then test the computer model to predict its future performance. For example, they could use it to predict how much stress the column could handle before it fractured under pressure.
How It Works
The AI was developed by taking MRI and micro-CT scans of vertebrae from a cadaver to generate realistic micro-structural models of the spine. They were then able to simulate what might happen if cancer weakened some of the vertebrae to understand how that would affect how much stress the bones could handle before fracturing. The model helped predict how much strength parts of the vertebra would lose as a result of the presence of metastatic spinal tumors, as well as other structural chances that could be expected as the cancer progressed. They then compared their projections to clinical observations from cancer patients.
Researchers say the model and the idea of a digital twin can help spine specialists understand new therapies, simulate different surgical scenarios and better comprehend how spinal vertebrae will change over time due to the presence of the tumor or the effects of radiation. The team hopes that in the future, digital twins can be hyper-specific to the patient.
“The ultimate goal is to develop a digital twin of everything a surgeon may operate on,” said Soheil Soghrati, co-author of the study and associate professor of mechanical and aerospace engineering at Ohio State University. “Right now, they’re only used for very, very challenging surgeries, but we want to help run those simulations and tune those parameters even more.”
As it stands, ReconGAN was developed using only one specific cadaver sample, and more data is needed to better perfect the technology. That said, even in its infancy the project is helping surgeons learn more about spinal issues and how to best treat them.
If you need assistance overcoming a spinal issue of your own, or you just want to get to the bottom of your back pain, give Dr. Sinicropi and the team at The Midwest Spine & Brain Institute a call today at (651) 430-3800.
