HISTORY OF CBMT
The evolution of MRTA into OsteoDx’s Cortical Bone Mechanics Technology™
Mechanical Response Tissue Analysis
CBMT is comprised of certain improvements to a prior technology called Mechanical Response Tissue Analysis (MRTA). Included in MRTA is the application of static and oscillating forces to the forearm, the measurement of the associated force and acceleration data, and the fitting of frequency response functions derived from that data to a mathematical model of the mechanical skin-bone system of the forearm.
MRTA was invented by Charles Steele, a professor in the Mechanical Engineering Department at Stanford University, in response to an expressed need of NASA for a means to assess the skeletal health of astronauts after space flight. Stanford received a patent on the technology in 1991 and licensed it to a small business, Gaitscan, Inc. for commercialization.
During the 1990s, Gaitscan built several MRTA systems with funding from NASA, and these were extensively utilized in animal and clinical research. (See the list of MRTA publications on the Resources page.) Meanwhile, Dual-energy X-ray Absorptiometry (DXA) was endorsed by the World Health Organization and that technology captured the market for diagnosing osteoporosis. As a result, Gaitscan was unable to raise the capital needed to continue developing the technology. Gaitscan shut its doors and Stanford abandoned its patent in 1999.
NASA took possession of the MRTA systems that Gaitscan had built and distributed them to investigators at several universities who continued to use them in their research. Without technical support, these systems could not be repaired or improved, however, and they all eventually fell into disuse.
Along the way, MRTA systems also acquired a bad reputation for measurement irreproducibility. Repeated measurements of the same person made within minutes of one another sometimes varied by as much as 65%. This reputation reduced the willingness of research sponsors to fund further use of the technology.
The reputation of the technology was worsened, we believe, by a strategic error in its development. Before he had mastered the application of MRTA technology to the nearly ideal biomechanics of the ulna bone in the forearm, Charles Steele succumbed to pressure from clinical investigators to try to apply the technology to the much more biomechanically complex tibia bone in the foreleg. This led to several years of unsuccessful effort while opportunities to master the application of MRTA to the ulna languished.
Development of Cortical Bone Mechanics Technology
In 2009, Professor Anne Loucks of the Biological Sciences Department at Ohio University read some papers about studies in which MRTA technology had measured effects of exercise training on ulna and tibia stiffness (Miller07, Miller09), and effects of KOH immersion on emu tibia stiffness (Wynnyckyj09). After discovering that the technology was not commercially available, and that it had acquired a bad reputation for measurement reproducibility, she resolved to find out why it didn’t work. She assembled a team of electrical and mechanical engineers to develop a better MRTA system, and recruited a steady supply of undergraduate biology students as technicians.
Thus ensued several years in which Anne’s team discovered and corrected several sources of error in MRTA data collection and analysis. One source of error in MRTA derived from the high sensitivity of measurement results to probe positioning on the forearm. This source of error is compounded by the inability of a MRTA or CBMT operator to recognize by sight or touch exactly where to put the probe. The correct location is obscured by the soft tissue covering the ulna bone, by the large differences between the shapes of ulna bones from person to person, and by large inter-personal differences in the chaotic pattern of destruction of cortical bone in the ulna caused by bone resorption ― which begins in early adulthood.
Another source of error derived from the non-unique results of MRTA data analysis. Alternative representations (such as the obverse and inverse) of the same accelerance frequency response function (FRF) data set led to different results. Results also varied with the frequency range over which a particular FRF was fitted to the mathematical model of the mechanical skin-bone system of the forearm, and MRTA offered no non-arbitrary reason to prefer one frequency range over another. Improvements correcting these and other sources of error became the subjects of US and international patent applications filed by Ohio University, and it is these improvements that comprise the novelty in CBMT.