Artificial intelligence transforms the modeling of the lumbar column for faster care to the patient

Almost 3 out of 10 adults in the United States have experienced lower back pain in any three -month period, so it is the most common musculoskeletal pain. Back pain remains one of the main causes of disability worldwide, affecting millions and, often, leads to chronic discomfort, lost work and invasive procedures.

Researchers and doctors are increasingly resorting to the modeling of the lumbar column, which unites engineering and medicine, creating a virtual and specific model of the patient from the lower back. This technology simulates how the column moves, where mechanical stress accumulates and what could be causing pain or dysfunction.

These detailed models are used to plan surgeries, evaluate spinal implants and develop personalized treatment strategies adapted to the anatomy of each patient. Despite its promise, the current modeling of the lumbar column is slow, manual and demands specialized experience, limiting scalability and personalization. This hinders clinical application and results in inconsistent results.

Researchers at the Florida Atlantic University Faculty and Informatics Faculty and the Marcus Neuroscience Institute in the regional mouth hospital, part of Baptist Health, have reached an important milestone in the modeling of the lumbar column by integrating artificial intelligence with biomechanics to transform the column diagnoses and personalized treatment planning.

They are the first to create a completely automated finite elements analysis pipe specifically for the modeling of the lumbar column. Its advance implies the integration of deep learning tools such as Nnunet and Monai with biomechanical simulators such as Gibbon and Febio.

The results of the study, published in the magazine World Neurosurgery, show that this new approach reduced the preparation time of the lumbar column model in 97.9%, from more than 24 hours to only 30 minutes and 49 seconds, without compromising biomechanical precision. The fully automated pipe allows rapid and specific simulations of the patient that support preoperative planning, spinal implant optimization and early detection of the degenerative conditions of the column.

The tests showed that the virtual column reacted as a real, with realistic movement of the disk, ligament tension and pressure on the back of the column during flexion and stretching. Because the system works with very little manual work, it is much faster and more consistent than traditional methods, which makes it a valuable tool for doctors and researchers equally.

What distinguishes our approach is its ability to automatically convert standard medical images such as computerized tomography or magnetic resonance in highly precise and specific spine models of the patient. Traditional manual methods require complex geometry processing, mesh and simulation configuration of finite elements, which makes them not only intensive in time but also highly dependent on the operator’s experience. Our automated pipe significantly reduces the required time, reducing what once took several hours or even days up to only minutes. “

Maohua Lin, Ph.D., corresponding author and Assistant Research Professor, FAU Biomedical Engineering Department

For the study, the researchers used Advanced to automatically identify important parts of the spine, such as bones and discs, medical scanning. Then they became smooth 3D models that included bones, cartilage and ligaments. They map where the ligaments bind and shape the cartilage depending on the common patterns. Finally, the researchers conducted computer simulations to see how the column respond to movements such as flexion and twist, helping them to understand where stress accumulates and how the column moves in real life.

“Beyond advancing in the investigation, the automated modeling of the lumbar column plays a fundamental role in preoperative planning,” said Frank D. Vrionis, MD, corresponding author and head of neurosurgery at the Marcus Neuroscience Institute. “This technology quickly generates specific patient models to predict mechanical complications, optimize implant design and reduce surgical risks. By eliminating manual steps, it also improves speed and consistency, helping doctors to make more informed decisions.”

This research is based on previous works of the research team published in leading magazines that include the review of artificial intelligence and the North American Spine Society Journal, which investigates biomechanical modeling techniques based on related AIs.

“This innovative work exemplifies the power that changes the game of unite engineering and medicine to address the complex challenges of medical care,” said Stella Batalama, Ph.D., Dean of the FAU Faculty of Engineering and Informatics. “Fau and Bautista’s health researchers are not only overcoming the limits of innovation, but also delivering real world solutions that can improve patients’ results and redefine column care.”

The co -authors of the study are Mohsen Ahmadi, Ph.D. student in the Department of Electrical and Informatics Engineering of FAU; Xuanzong Zhang, a student of the Highlight High School; Yufei Tang, Ph.D., Associate Professor, Department of Electrical Engineering of FAU and Institute of Computer Science and FAU Sensing Institute; Erik Engberg, Ph.D., Professor, Department of Biomedical Engineering and Department of Mechanical Engineering of Ocean and Mechanical, member of the Fou center for complex systems and brain sciences within the Faculty of Sciences of Charles E. Schmidt, and member of the Fau Stiles-Nicholson Brain Institute; and Javad Hashemi, Ph.D., inaugural president and professor of the Department of Biomedical Engineering and Dean of Research, Fau College of Engineering and Computer Science.

This investigation was supported by Boca Mouse Regional Hospital, part of Baptist Health, the Helene and Stephen Weicholz Foundation, the National Foundation of Sciences, the Pilot Subsidies of the Fau College of Engineering and Computer Science, the FAU Stiles-Nicholson Brain Institute, the FAU Center for Smart Health and the Institute of FAU.