October 9, 2023 – Artificial intelligence tools for treating hip fractures are on the rise.
The latest study from researchers on the University of Pennsylvania can predict a patient's risk of dying after a hip fracture, an underappreciated health threat that affects lots of of 1000's of Americans annually.
The innovation could help doctors flag high-risk patients so that they might help and potentially save lives.
“We wanted to try a bunch of different AI algorithms, feed them all the information, and then figure out what is the most accurate predictor of mortality that we can get,” said study co-author Abhinav Suri, a medical student on the University of California, Los Angeles.
Other recent advances can predict the chance of a hip substitute fracture as well recreation mobility after hip surgery. Previous efforts also used AI to evaluate death risk after Hip fracturehowever the new study According to the researchers, more algorithms were tested and more patient data were included.
Researchers used a decade's value of information from 3,751 hip fracture patients to coach ten machine learning algorithms. The resulting models provide a “mortality risk score.” Models were assessed on how well they predicted mortality at 1, 5, and 10 years after hip fracture.
The models “learned” from the outcomes of 149 laboratory tests and 7 demographic variables. From these data points, the researchers identified ten characteristics which might be most vital for mortality risk. At the highest of the list was age, followed by blood sugar levels.
“There is currently no mortality risk calculator for hip fractures,” said Cory Calendine, MD, an orthopedic surgeon on the Bone and Joint Institute of Tennessee who was not involved within the study. Some methods, similar to the Charlson comorbidity index, might help predict death more globally, but have limited utility in fracture care.
The model may not change the way in which doctors treat hip fractures, which nearly at all times require surgery. But it could help doctors advise the family or signal a medical examiner to recommend more frequent or intensive follow-up care.
How AI could help tackle a “tremendous public health problem.”
More than 300,000 Every yr people within the United States break a hip. Below, 20% to 40% die inside a yr, and a 3rd of those that survive longer lose theirs independence.
“Hip fractures are a huge public health problem. “The truth is, it really requires a preventive approach,” said Cody C. Wyles, MD, assistant professor of orthopedic surgery and clinical anatomy on the Mayo Clinic, who was not involved within the study. “One of the reasons the injury is so devastating is not so much the injury, but rather that it is a sign of poor health.”
According to Wyles, many patients with hip fractures have poor bone mineral density, muscle strength and immune function. For these patients, not having the ability to move around after surgery might be devastating.
According to Wyles, organ failure shortly after hip substitute surgery is “very rare” but might be attributable to the discharge of bone marrow into the body, each when the fracture occurs and after the location of bone implants during surgery. Bone marrow can flow into into the lungs, putting strain on the guts and blood vessels.
Wyles led the Orthopedic Surgery Artificial Intelligence Laboratory at Mayo Clinic, where researchers develop risk prediction tools for patients and use AI to create synthetic versions of patient X-rays that might be viewed from multiple angles. AI models may determine targets for robotic tools to aim at during an operation.
But with regards to hip fractures, “nutrition and exercise will be far more important than AI in helping us address this crisis,” Wyles said.