‘Laboratory For Motion’ Uses AI To Turn Movements Into Clinical Endpoints
By Deborah Borfitz
August 25, 2022 | Movement can now be used as a functional biomarker and endpoint for all sorts of treatments—including orthopedic surgery for bad knees and hips, spinal cord stimulation for debilitating back pain, weight-loss programs for obesity, and even laser surgery for glaucoma. Physicians are really all rehabilitative medicine specialists whose goal is to get patients to move better, according to Frank Fornari, Ph.D., chairman and founder of BioMech Holdings, LLC, which has created a clinically relevant “laboratory for motion.”
Fornari has been doing mathematical modeling work since the 1980s, but only recently has the hardware and processing power become available for the initial artificial intelligence (AI)-based development of motion analytics applications in all its dimensional complexities in a matter of weeks rather than years, he says. Any aspect of movement that clinicians might want to quantify relative to a norm can be easily captured in minutes or seconds by the BioMech Lab platform.
The alternative is often a combination of subjective opinion and medical guesswork, says Fornari. That helps explain why so many people get surgeries that don’t help, or were never needed, and may stay on pain medications longer than necessary.
“BioMech has not only developed tests that analyze integral aspects of motion including range of motion, gait analysis, balance, symmetry, and cognition,” Fornari says. “We have also developed tests written for each specific motion because they are standardized across healthcare verticals.”
As opposed to consumer-based fitness trackers, this is a “serious medical platform with serious computing power... configured in a unique way to increase the signal-to-noise ratio,” says Fornari. “We have used innovative engineering to make sure that signal is precise, accurate, and reproducible.”
BioMech is revolutionizing the way medicine gets delivered, says Fornari, who puts the platform technology in the same league as the first MRI. “It is a pretty big step forward.”
Practical Applications
BioMech Lab launched only a few years ago and is already in use around the world used by a variety of clinical specialties, Fornari says. The technology received the AI Breakthrough Award for “Best Overall Biometric Solution” after it was utilized for follow-up functional mobility testing on individuals with Parkinson’s disease in a non-contact boxing program designed to ease their symptoms. The award recognizes BioMech Lab for its use of motion as a functional biomarker to quantify relevant aspects of physical, surgical, pharmaco-, and cognitive therapies.
The neurosurgery department at Virginia Commonwealth University (VCU), Fornari’s alma mater, has been using the platform for several years in a clinical trial to support the diagnosis and treatment of normal pressure hydrocephalus (NPH)—a condition that mimics hydrocephaly but without the increased pressure on the brain—in lieu of observing their gait, balance, and cognition right after a spinal tap. In the past, he notes, patients would have been asked to walk down the hall followed by a team of neurosurgeons, physician assistants, and nurses to see if they detected any differences. “It was inefficient, and it is the standard.”
Diagnostically qualified individuals get surgery to place a shunt into their brain to drain the excess fluid, he continues. Pressure settings get adjusted using functional motion metrics provided by BioMech Lab rather than, as previously, relying on input from watchful family members.
“We’ve completely changed the way [neurosurgeons can] do things,” says Fornari, drawing a comparison with titrating chemotherapy using white blood cell count as a biomarker rather than measuring the gene or receptor that the compound works on. “Now, they can use BioMech Lab to determine how to adjust the shunt and get patients their function back.”
A clinical trial underway by orthopedic surgeons at the Duke University School of Medicine is using the platform for before-and-after gait evaluations in patients getting a hip or knee replacement. The lack of quantitative assessments to determine when a new knee joint is needed is behind the high failure rate (35%) against which payers now exact financial penalties, Fornari says.
“As a result of the work at Duke, we were able to demonstrate a strong correlation between the clinician’s assessment of disease state and BioMech Lab’s functional analyses,” he adds. “These data are a very positive step toward creating a much more objective patient assessment tool.”
Tackling Back Pain
At Integrated Pain Solutions (Columbus, Ohio), medical director Gladstone C. McDowell II, M.D., is partnering with BioMech in a trial demonstrating the utility of spinal cord stimulation in individuals experiencing incapacitating back pain. Specifically, the study is measuring functional motion and cognition in patients to see if they improved enough to justify use of a permanent stimulation device (Senza, Nevro Corporation).
The study, funded by Nevro and approved by a centralized institutional review board (WCG IRB), initially involved 25 adult patients who were implanted with the Senza Spinal Cord Stimulator over the past year, says McDowell. Improvement to their gait and balance were assessed prior to trial; one week after trial; and four, eight-, and 12-weeks post-implantation.
McDowell says he was thrilled to learn about the BioMech Lab platform because “there is no blood test to tell whether someone is in pain or not or how much pain they are in, [and] there is no EKG or scan [the neurogenetics and biochemistry are too complicated, Fornari points out]. It comes down to whatever they tell us, and we all know people lie about their age and lie about their weight, and some of them lie about pain.
“Patients may not want to tell us they’re feeling better because they’re afraid we’ll cut back on their pain medications, and they may not get their disability or their time off,” McDowell continues. “On the converse we need to know when people are functioning better because... pain is a subjective thing that varies person to person, so we like to combine that with function,... [like] are you sleeping better, are you walking better, are you stronger, are you falling less or able to stabilize yourself more.”
The Nevro pacemaker-type device, which attaches to stimulator wires tunneled underneath the skin, would be implanted at a surgery center, explains McDowell. Patients are given a remote control that they can turn on or off to adjust therapy settings.
“Spinal cord stimulation is basically using electrical pacemaker wires inside the spine to block the sensation of pain and in some ways inhibit the brain from experiencing that pain,” McDowell continues. The technology has been around since the 1960s, enhanced by Medtronic in 1981 when it came up with the first implantable battery.
Nevro further advanced the field with algorithms and technology that significantly sped up spinal cord stimulation frequencies from a ceiling of about 1,200 hertz up to 10,000 hertz where the vibrations couldn’t be felt, he says. The use of spinal cord stimulation skyrocketed after the device gained the approval of the U.S. Food and Drug Administration in 2015, from a growth pace previously sitting between 3% and 4% per year to an almost 19% spurt in each of its first two years on the market.
McDowell’s objective with the clinical trial was to quantify if and how well his patients were doing, in terms of their functioning and not just how they said they felt, after spinal cord stimulation with the Senza device, says McDowell. The goalpost for moving patients on to the permanent implant was 50% or more relief from dysfunction.
Encouraged by findings in the first 25 patients, and with additional financial support from Nevro, the study is ongoing and will allow for the collection of long-term data on the first group of patients who had the device implanted last year, McDowell says. Findings from the first three months were presented at the recent American Society of Pain & Neuroscience meeting and the abstract won best-of honors among hundreds of submissions.
Molecules to Organisms
Fornari himself has a vast and varied background that early on included directing clinical trials worldwide for a pharmaceutical company. He earned a master’s degree in neurophysiology, attended medical school, and has a Ph.D. in molecular pharmacology and toxicology as well as postdoctoral training in synthetic chemistry and drug design. While in academia, he helped develop a compound that launched a pharmaceutical venture.
Following the pharmaceutical venture, he and his wife Gwen Bauer, Ph.D., chief science officer at BioMech, co-founded the country’s largest clinical pharmacological monitoring laboratory, Dominion Diagnostics, where they developed new technology that permitted the rapid identification and quantitation of drugs and their metabolites, he shares.
Clinicians use quantitative urinalysis to see if their patients are metabolizing a drug properly based on their elimination rate, which is a mathematical constant, Fornari says. “They just take a sample, and it tells you exactly what drug or drugs are in the patient, what metabolites are present, and how they are eliminating the compound(s), removing the need for subjective patient self-reporting and guesswork.”
Fornari was doing mathematical modeling in the 1980s and later, in collaboration with colleagues at VCU Medical Center (then called Medical College of Virginia), they put together one of the country’s first molecular modeling laboratory for designing compounds on a computer screen using algorithms for motion.
Mathematical analysis is at the heart of everything Fornari has ever done. Given his success with analyzing motion at the molecular level, it made sense that it should also be possible to build a system to do the same thing at the organismal level, he says.
The ability to move is, after all, intrinsically tied to health. “Whether you’re an oncologist or an orthopedist, it doesn’t matter,” Fornari says. “You want patients to get up, get on with their life, go up and down stairs, go play tennis, do what they have to do, and disease stops that or affects those functions.”
At-Home Possibilities
“The platform is suited to analyze anything that affects motion, good or bad,” says Fornari, citing the example of a geriatric patient who refused to go to physical therapy after getting new knees but will stand in front of a television screen competing for the best balance score. “It focuses her away from the pain and on her performance.” The surprising reality is that when patients can see that they’ve regained movement and balance, they tend to better tolerate pain.
BioMech is currently in discussions with pharmaceutical companies who want to use the motion analysis platform in their preclinical studies and during clinical trials to assess the efficacy of a drug for a movement disorder or to quantify the side effects of a drug. A national home health company has also reached out, adds Fornari, noting that the Centers for Medicare and Medicaid Services has created three new codes allowing these devices to be a reimbursable expense when used in the home.
As more and more healthcare services get delivered at home, the hospital of the future is going to become a “catchment area” for diagnostic vendors—including MRI companies like Siemens and clinical laboratories like Quest Diagnostics as well as motion laboratories like BioMech Lab—who get paid the same lower rate as they would be anywhere else, he predicts. “And it is going to give clinicians more time to do the things they are trained to do, give patients more access to comprehensive healthcare, and result in large savings for the healthcare system.
“Payers are looking for companies like ours that are coming up with mobile and portable solutions,” says Fornari. “We are building products for home use right now.”