The primary mission of the SPPDG continues to be electronics research in pursuit of objectives specified by the sponsoring organizations. Delivery on the contracts, and excellence in both research and support to the sponsors, are the keys that have made the SPPDG successful thus far.
There are, however, opportunities for technology transfer between the leading edge of the electronics industries represented by SPPDG sponsors and industrial collaborators, and the direct patient care medical community represented by the Mayo Foundation clinics and hospitals. These opportunities arise only when members of one community interact with the other, as occurs routinely in the SPPDG. Since 1989 the SPPDG has participated in approximately thirty such technology transfer projects with other staff members of the Mayo Foundation. Although typically several such projects are active at any one time, a large prior project and a set of presently developing projects are noted below as examples.
Global Positioning System/Moving Map Display in Mayo’s Medical Evacuation Helicopters
In 1988, our group, teamed with an Israeli software development group, developed and installed the first example of a helicopter navigation system using a global positioning system (GPS) receiver and a moving map display on a laptop computer installed in “Mayo One”, Mayo’s medical helicopter. This project continued until the late 1990s, at which point several commercial companies mass-produced such systems for installation in a wide variety of medical helicopters. In our early 1990s studies, we were able to demonstrate that, at the scene of, e.g., an auto or farm accident in a rural area, if the GPS coordinates of the accident were supplied to the helicopter crew, the helicopter could fly directly to the accident scene. The round-trip time savings were typically roughly twenty minutes, or one-third of the “golden hour” that is critical to life-saving efforts for injured patients.
NASA Advanced Communications Technology Satellite
NASA developed, launched (in 1993), and tested an Advanced Communications Technology Satellite (ACTS) to demonstrate the effectiveness of new antenna concepts and on-board digital processing of data streams.
The SPPDG was contacted by NASA in 1992 to ascertain the level of possible interest at Mayo Foundation in participating in the ACTS experiments, and the original proposals to NASA and DARPA for funding for the clinical experiments was prepared by the SPPDG. The Mayo Foundation participated in the ACTS experiments program to investigate communication techniques which may eventually allow large medical centers to provide supporting medical services to small and medium-sized medical facilities in small towns and rural areas. While at that time Mayo had a decade of experience in the use of conventional analog video satellite communications for "telemedicine" between our three primary sites in Rochester, MN, Scottsdale, AZ, and Jacksonville, FL, the Phase One ACTS experiment, conducted at 1.5 megabit/second data rates, was intended to demonstrate that the provision of quality medical diagnostic and information services to remote facilities can be cost effective and timely. These experiments were concluded successfully, with the result that several research papers describing the clinical efficacy of the Phase One experiments were published in the Mayo Clinic Proceedings.
The Phase One experiments were followed with a Phase Two set of clinical demonstrations, in which much higher data rates, i.e., 155 megabits/sec (the Asynchronous Transfer Mode [ATM] protocol referred to as STS-3) were employed. For these experiments data in ATM format was transmitted by fiber optic landline from Rochester, MN, to Kansas City, KS, then uplinked to the ACTS satellite, and received by a specially designed ACTS earth station located on the premises of the Mayo Clinic Arizona in Scottsdale, AZ. Nearly a half dozen different types of telemedicine and clinical outreach experiments were conducted. The results of these studies were published in four papers in the August 1999 Mayo Clinic Proceedings and in IEEE Network (see References 187 and 188 on the Publications page). The information gained from these studies was employed by Mayo Foundation in the design and implementation of a next-generation video communications system linking our three primary clinical care sites in Rochester, MN, Jacksonville, FL, and Scottsdale, AZ; the system employs the ATM data transmission protocol, with the encrypted data transmitted on fiber optic landlines through the so-called "public switched ATM network" provided by the commercial telecommunications industry.
Technology Transfer to the Electronics Industry
The SPPDG works closely with the Mayo Clinic group referred to as Mayo Clinic Ventures (MCV; the commercialization arm of Mayo Clinic) to license the rights to technological innovations, where possible, to commercial companies. SPPDG staff members have prepared several dozen technology disclosures, on a wide variety of electronics, optics, and micro-electromechanical systems (MEMS) concepts and devices, as well as medical devices (see below) to MCV. A number of these disclosures have resulted in issued patents, while others are working their way through the U.S. patent process. MCV is constantly increasing its contacts with the commercial electronics industry, and welcomes queries from that industry regarding SPPDG inventions and developments.
Small Autonomous Body-Worn Physiological Monitors
The SPPDG constantly examines possible areas for technology transfer of electronics-based devices and systems into Mayo's clinical practice. We have been working since the early 2000s with various Mayo clinical collaborators to develop a family of very compact body-worn physiological and analyte monitors intended, not for consumer self-help applications, but rather for clinical applications. The intent is that the units that we have developed and are developing will produce data of sufficient quality that, once converted into information, can be used to inform clinical decisions regarding Mayo’s patients. The units are intended to be employed in both clinical research and to be prescribed by physicians to their patients.
Beginning in 2003, in a collaboration with a member of the clinical staff of Mayo's Division of Endocrinology, Diabetes, Metabolism, and Nutrition, we developed the first version of a self-contained body-worn physiological monitoring unit to measure activity in obese patients being treated at Mayo Clinic. This early unit was referred to as a Posture and Activity Detector (PAD). Even at that time the unit contained a 3-axis accelerometer, a temperature measurement unit, and a nonvolatile memory, which could store data continuously for up to a month. These early units, crude by present-day standards, were used in preliminary studies of healthy and normal obese adults at Mayo, and in obese and lean children (in collaboration with researchers at National Institutes of Health). Descriptions of the initial work on normal adults and children can be found in Number 250 in the Publications list on this web site. The photo below depicts an example of these early units, without a case, and with two different experimental cases intended to investigate the physical robustness of the devices.
Posture and activity detector 1 (PAD1) printed circuit board assembly shown next to two different styles of PAD1 case assemblies. Outer dimensions of the PAD units in their cases are 3.2" (82.75 mm) long, 1.5" (38 mm) wide, and .54" (13.75 mm) thick.
All of these units are viewed as members of a "platform technology", in that, by adding software and hardware features, a number of measurement functions can be incorporated onto the same basic “platform”. In their most recent embodiments (e.g., since 2010) we have demonstrated accelerometer-based data recording units, accelerometer-plus-ECG recording units, and accelerometer-plus-ECG-plus-pulse oximetry units. The newest motion-only recording units have been used by a research team in Mayo’s Department of Orthopedics to assess gait and mobility in lower limb-compromised patients and in volunteers, resulting in a half-dozen published papers in orthopedics journals.
The photo below depicts two different units, developed over the past five years, with different form factors and different levels of functionality. The unit on the right is our most recent motion-only recording unit, containing three three-axis accelerometers, a microcontroller, a 1 GB flash memory, a battery (allowing two-week 24/7 recording), an accurate time-of-day clock to allow precise time-stamping of the recorded data, and other assorted circuitry. The unit on the left, assembled on the same basic platform, contains, in addition, a larger battery and high resolution low-power ECG monitoring circuitry (one lead pair, 1000 samples/second, 12 bits/sample), also with two-week recording capacity. The larger unit was developed for an expedition by professional climbers to Mount Everest in the Spring of 2012, sponsored by the North Face apparel company, Mayo Clinic, and National Geographic. Of the ten climbers who reached the summit, eight were wearing the Mayo devices. Upon the return of the devices to our laboratory at the end of the expedition, we were able to extract 95 GB of high quality motion-and-ECG data from all of the climbers who wore the units.
Two versions of body-worn physiological monitoring unit developed by this research and development group. Leftmost unit is ruggedized for use in physically harsh environments by, e.g., mountain climbers. The unit supports single-lead high resolution ECG (1000 samples/second) and three 3-axis motion monitoring units. All data is recorded on the units. Run time is 14 days, 24/7. Rightmost unit is a motion-only recording unit for use with volunteers in clinical research studies, and for selected patients. Run time is 14 days, 24/7.
The last photo below depicts one of the single lead-pair units monitors on the Mount Everest expedition, placed on a hand to illustrate the size of the units.
Another view of the ruggedized body-worn physiological monitoring unit depicted in the previous figure, illustrating the two ECG electrode wires. A three-wire unit has also been developed, but not fielded into harsh-environment usage.
In addition, at present we are refining the pulse oximetry circuitry, and working on the further addition of a low-power impedance-based respiration monitoring capability to the units as well. We believe that these additional functions, and more, can be incorporated into a footprint the same size as, or slightly smaller than, the units that were deployed to Mount Everest. For further information on these body-worn units, please refer to Numbers 311 and 312 in the Publications list on this web site.
Wireless Versions Of The Body-Worn Units, And Their Incorporation Into An End-To-End Home Monitoring System
We have developed wireless versions of the body-worn physiological monitoring unit, as illustrated in the following photo. The wireless body-worn unit is the first element in an end-to-end wireless-based home health monitoring system with bidirectional communications capability between the wearer’s home environment and a medical center. In the leftmost view of the photo below, the meander-line antenna is clearly visible; the rightmost view shows the 300 mAh thin film battery that powers the unit. The unit itself has all of the same sensing and data storage functionality of the previously described units, with the addition of the wireless function.
Front and back side views of a version of the body-worn unit with short-haul wireless transmission capability; the device supports three 3-axis accelerometers as well. The antenna is a specially designed meander line. U.S. quarter (23 mm diameter) references the size of the unit; note also the 300 mAh label on the thin-film battery.
The body-worn unit in the photo depicted above incorporates an antenna of our own design. Although several published manuscripts that describe wireless-capable body-worn physiological monitors have employed commercially available inexpensive 2.4 GHz antennas (e.g., chip antennas and similar devices), we wanted to evaluate radiation patterns from various physically and electrically small antennas that would function most efficiently in a close-to-the-body location. Antenna radiation patterns are typically influenced by their surroundings, and in the case of commercial chip antennas, the ground plane structures in the circuit boards on which they are mounted become a primary component that shapes the antenna’s radiation pattern. It was unclear to us that chip-type antennas would have the greatest performance in the body-worn environment that could be achieved.
As part of a detailed study, conducted without any preconceptions, of antenna designs that might have improved radiation and matching characteristics for use with body-worn units, we conducted a broad set of electromagnetic simulation and modeling tasks, which resulted in a different approach to the configurations of antennas for these applications. Based on the results of the simulations, we then designed a half-dozen antennas, using a meander-line configuration, and iterated these designs several times. The antennas were fabricated and tested in our radio frequency anechoic chamber (see the “Test and Measurement” section of this web site) in conditions as close to real-world as we could achieve. The antennas were designed to operate in the 900 MHz band, which the simulations demonstrated is more efficacious than the 2.4 GHz band for these body-worn applications. The following figure depicts one of these sets of six different antenna designs. For further information, please see manuscript #314 in the Publications section of this web site.
Printed meander line antennas for short haul wireless link; six antenna designs printed on 0.006" thick printed circuit board dielectric material. Each design includes multiple shorting pin and feed network connections for fine tuning.
The complete end-to-end monitoring system is illustrated diagrammatically in the next figure. The body-worn unit, illustrated in cartoon format attached to the wearer at the waist (though the unit is positioned on the body as needed to support its monitoring function), transmits a very low power radio frequency (RF) signal to a wireless gateway up to thirty meters distant, located within the wearer’s residence This unit is not a WiFi gateway, but one specifically designed to support the body-worn unit’s functionality. The gateway receives and then converts the wirelessly transmitted packets to Ethernet packets, which can then be fed into any of four different long-haul transmission paths, as illustrated in the figure below. Of these four possible pathways, we have successfully tested the terrestrial wireless path using both the WiMax and LTE protocols; and the cable modem path. The terrestrial wireless path was accomplished at a carrier frequency of 3.65 GHz, using a license provided to us by the U.S. Federal Communications Commission. A test of the satellite link was only partially successful. A redesign of the timing chain in the gateway will be needed to account for the long round-trip delay between the gateway and a satellite in high earth orbit, but is practical to achieve.
Illustration of several different approaches to full duplex communication between a wireless body-worn physiological monitoring unit and a medical center. The contents of the oval illustrate those portions of the system located within the home. The notional home illustrates a rooftop-mounted antenna, but in actual practice the antenna is much smaller, and can be located in an attic or on a high shelf in an apartment. The branch exchange illustrates the link from the home if cable or DSL modems are used; the large building on the right side of the diagram represents a medical center building.
Using a version of the wireless unit containing both motion monitoring and ECG monitoring capabilities, we programmed the body-worn unit’s firmware to transmit a snapshot of ECG data, a five-second burst out of every thirty seconds of elapsed time. A section of this data transmission appears in the following figure.
ECG waveform digitized and transmitted from a physiological monitoring unit on a volunteer, the unit was programmed to transmit five seconds of data out of every thirty seconds. Upper panel: a 375 second data snapshot. Lower Panel: a time-expanded view of one of the five-second “snapshots”. The WIMAX long-haul transmission protocol was employed, at an FCC-assigned transmission frequency of 3.65 GHz.
We demonstrated the ability for the wireless body-worn unit to contact the in-home gateway at very low transmitted RF power levels. The next figure illustrates the results of radio frequency measurements made in the home of a volunteer. At a transmitter power of slightly less than 1 mW, it was possible to “close the link” between the body-worn unit and the gateway at every location tested throughout the dwelling, including when the wearer was lying face-down on top of the unit.
Test of the wireless physiological monitoring unit and the home gateway in the dwelling of a volunteer, illustrating that RF link closure was established in every location tested, between the body-worn unit and a centrally located gateway. Power output of the transmitter in the body-worn unit was slightly less than 1 mW in the 900 MHz band.
Additional physiological and biochemical analyte monitoring capabilities are being added to subsequent generations of the body-worn unit. For more information regarding this wireless-based monitoring capability, please refer to Number 313 in the Publications section of this web site.