NOTE: End date changed to 9/30/2013 per NSBRI data submission (Ed., 3/4/2014)

One of the key questions that remains unanswered as we prepare to send humans to other planetary surfaces is the degree to which living and exercising in these reduced gravity environments will provide an osteoprotective stimulus to prevent the loss of bone mineral density (BMD) that has been observed in microgravity. The concept of daily load stimulus is useful in this regard since it has the potential to estimate the "dose" of load to the lower extremities that will maintain skeletal integrity even in the setting of concurrent therapeutic drug and exercise countermeasures. Most observers believe that some form of supplementary exercise will be required activity on the moon, Mars, or nearby asteroid, but this will need to be optimized to provide the most efficient use of crew time. Cavanagh et al. (J. Biomech., 2010) have published reports that, on average, only 43 minutes of the ~150 minutes assigned for exercise during a day resulted in loaded exercise. Given the continued loss of BMD observed in crew members after long-duration flights, this amount of loaded exercise is likely not enough to preserve an acceptable amount of bone strength.

The Specific Aims of the project were to:

1) Develop hardware based on Micro Electro Mechanical Systems (MEMS) technology that can unobtrusively monitor the accelerations applied to the body and interface with an ambulatory monitor.

2) Extend the Daily Load Stimulus Algorithm to account for recent developments in bone mechanobiology, to incorporate accelerometric signals, and to write software to perform this analysis in real-time.

3) Demonstrate the feasibility and validity of the approach in 1g, in 1/6g in the eZLS, and in another analog.

4) Integrate the hardware and software into a package suitable for flight development.

2. Key findings to date

• Foot forces suggest IVA lunar and Martian locomotion (walking, running, loping, hopping) alone may not be osteoprotective, and that a simulated lunar EVA (body weight plus ~200 Earth lbs of suit mass) locomotion may not be osteoprotective.

• Foot forces suggest that locomotion in a simulated Martian EVA (body weight plus ~200 Earth lbs of suit mass) may provide adequate loading under some locomotion conditions depending on the duration of the activities.

• Lunar and Martian hopping and loping result in higher foot forces than walking, while running provides the highest foot forces in any one environment.

• The Classification Trees developed can precisely recognize locomotor activity, which is useful during remote monitoring scenarios.

• Preliminary comparisons of tibial axial acceleration during parabolic flight to 1g confirm that tibial acceleration scales during the same activity in different gravity conditions.

• Preliminary results support the use of tibial acceleration as an indication of impact and therefore an effective tool for the examination of exercise regimens in reduced gravity.

3. Impact of key findings on hypotheses, technology requirements, objectives and specific aims of the original proposal

The wireless activity-tracking device has been designed, manufactured, and tested in a series of studies in 1g, 1/6g, and 3/8g simulated environments. Initial data analysis is starting to reveal potential crew health risks to bone health maintenance in reduced gravity environments. The sensor has been interfaced with a Smartphone to allow data collection in the field. This is an important step in moving towards flight readiness.

4. Future work

• Final parabolic flight campaign in April 2014.

• One of the next steps will be to repackage the system hardware into a final configuration including design considerations for deep space missions.

• Further enhance the Activity Recognition algorithms.

• Continue to test the interface between our wireless sensors and the BioNet software framework in the laboratory setting, for future automated data management aboard the International Space Station (ISS).

• Utilize the Smartphone platform as a portable data logger that is capable of communicating with the wireless sensors and the BioNet software framework.

• Review results from a data-sharing arrangement with the Integrated Medical Model team at NASA Glenn who received Jump Down data from this study, and with colleague Joern Rittweger who received static hopping data to assess the loss of energy during a stiff legged hop at 1g, 3/8g, and 1/6g.

• Prepare manuscripts for publication to peer reviewed journals.

The small wireless sensors developed in this study have been useful in other research efforts. Attached to the ankle, the sensor is currently being used in a study examining gait characteristics of female runners engaged in regular, long-distance training programs who are prone to injury. The goal is to reduce the rate of loading during running. A program has been developed to provide real-time feedback from the sensors as displayed on a screen in front of the runner. The runner can adjust their gait to a desirable data point in real-time, or take a portable data logger with them into the field and analyze the data post-run to see how the in-lab training has helped to re-train their running gait. Additional work has been proposed to use the sensors to assess the trade-off between hardware complexity and information density in comparing activity data from patients with bilateral osteoarthritis of the knee and unilateral trans-tibial amputations. Applications in the military are also envisaged where activity recognition in a large group of military personnel could be performed. We believe there are many other applications as well.

2) DEPLOY SOFTWARE UPDATES & INCREASE DATA THROUGHPUT: A comprehensive analysis of the wireless communication protocol and software configuration was conducted. Software updates have resulted in more reliable communication and more consistent data logging.

3) COMPLETE DATA COLLECTION IN PARABOLIC FLIGHT: Data was successfully collected on 9 subjects during parabolic flight campaigns in October 2012 and April 2013. However, neither campaign yielded a complete and successful data set. We are tentatively approved for another flight campaign in 2014 where we hope we can get a full complement of parabolas and activities/configurations with all data components. In depth analyses of the data collected is currently underway.

4) CONTINUED DATA ANALYSIS: Data analysis has been an on-going effort over the life of the project. Analysis of the data continues to suggest that the partial gravity environments of the moon or Mars will not alone be osteoprotective. Exercise will remain a necessary countermeasure in these environments. It is unclear what amount of loaded exercise is necessary to maintain optimal bone health. On-going work in this study will utilize the enhanced Daily Load Stimulus theory to help answer this question. It is certain that running in reduced gravity will still benefit from use of a subject load device to increase impact during locomotor exercises. Enhancements have been made to our activity programs to allow real-time reporting of daily activity and progress toward individual daily load stimulus target goals. NASA Glenn's Integrated Medical Model (IMM) team is exploring input of the jump down data into their bone fracture risk model, and with Joern Rittweger of the German Aerospace Association to explore the energy losses experienced during static, stiff-legged hopping in 1g, 3/8g, and 1/6g environments. We will be following their progress in analysis of these data.

5) FLIGHT READINESS: The system performed well during parabolic flight testing. The sensors have successfully been interfaced with the Smartphone platform which can be used as a portable data logging system. Currently one sensor communicates with the phone at a time, and the feasibility of connecting two Bluetooth devices to one phone is being explored.

One of the key questions that remains unanswered as we prepare to send humans to other planetary surfaces is the degree to which living and exercising in these reduced gravity environments will provide an osteoprotective stimulus to prevent the loss of bone mineral density (BMD) that has been observed in microgravity. The concept of daily load stimulus is useful in this regard since it has the potential to estimate the "dose" of load to the lower extremities that will maintain skeletal integrity even in the setting of concurrent therapeutic drug and exercise countermeasures. Most observers believe that some form of supplementary exercise will be required activity on the moon, Mars, or nearby asteroid, but this will need to be optimized to provide the most efficient use of crew time. Cavanagh et al. (J. Biomech., 2010) have recently published reports that, on average, only 43 minutes of the ~150 minutes assigned for exercise during a day resulted in loaded exercise. Given the continued loss of BMD observed in crew members after long-duration flights, this amount of loaded exercise is not enough to preserve an acceptable amount of bone strength.

The Specific Aims of the project include:

1) Develop hardware based on Micro Electro Mechanical Systems (MEMS) technology that can unobtrusively monitor the accelerations applied to the body and interface with an ambulatory monitor.

2) Extend the Daily Load Stimulus Algorithm to account for recent developments in bone mechanobiology, to incorporate accelerometric signals, and to write software to perform this analysis in real-time.

3) Demonstrate the feasibility and validity of the approach in 1g, in 1/6g in the eZLS, and in another analog.

4) Integrate the hardware and software into a package suitable for flight development.

2. Key findings to date

- Foot forces suggest IVA lunar and Martian locomotion (walking, running, loping, hopping) alone may not be osteoprotective, and that a simulated lunar EVA (body weight plus ~200 Earth lbs of suit mass) locomotion may not be osteoprotective.

- Foot forces suggest that locomotion in a simulated Martian EVA (body weight plus ~200 Earth lbs of suit mass) may provide adequate loading under some locomotion conditions to be osteoprotective depending on the duration of the activities.

- Lunar and Martian hopping and loping result in higher foot forces than walking, while running provides the highest foot forces in any one environment.

- The Artificial Neural Network (ANN) developed can precisely recognize lunar locomotor activity, which is useful during remote monitoring scenarios.

- An ankle mount configuration was tested in a small cohort (n=6). There is a marked decrease in signal magnitude between the ankle and in-shoe mounting. The ANN will need to be retrained with ankle mount data.

3. Impact of key findings on hypotheses, technology requirements, objectives, and specific aims of the original proposal

The wireless activity tracking device has been designed, manufactured, and tested in a series of studies in 1g, 1/6g, and 3/8g simulated environments. Initial data analysis is starting to reveal potential crew health risks to bone health maintenance in reduced gravity environments.

The sensor has been interfaced with a Smartphone to allow data collection in the field. This is an important step in moving towards flight readiness. The transition to the Smartphone interface has allowed a spin-off use of the wireless sensor system in a study aimed to reduce the rate of loading during running and ultimately reduce the rate of injury in female runners.

The NASA Human Research Program (HRP) has decided to not utilize the lunar bed rest model, and as such the system has not been tested in that analog as originally proposed. An alternative test environment such as a parabolic flight would be an ideal environment to test the system in, and the feasibility of conducting such a test is under review. The initial parabolic flight has demonstrated the system is ready for a full parabolic flight study. The study would provide the data needed to validate the system and compare to the results of the simulated reduced gravity environments test conducted in the eZLS. These tests would help to bring the activity sensor system to a high Technology Readiness Level (TRL=7). Final data analysis will help determine if additional software updates are necessary and help solidify drawings and design requirements that will be prepared for a CDR during Year 4.

4. Proposed research plan for the coming year

- Further enhance the Activity Recognition library to recognize activities collected in the ankle-mounted configuration and to include automated detection of additional mission critical tasks such as ladder climb, rock translation, platform jump down, squat exercise, and obstacle avoidance.

- Continue to test the interface between our wireless sensors and the BioNet software framework in the laboratory setting, for future automated data management aboard the International Space Station (ISS).

- Utilize the Smartphone platform as a portable data logger that is capable of communicating with the wireless sensors and the BioNet software framework.

- Test the activity monitoring system on a parabolic flight as a validation of system operation in a simulated Martian (1 parabola), lunar (2 parabolas), and microgravity (22 parabolas) environment.

- Review results from a data-sharing arrangement with the Integrated Medical Model team at NASA Glenn who received Jump Down data from this study, and with colleague Joern Rittweger who received static hopping data to assess the loss of energy during a stiff legged hop at 1g, 3/8g, and 1/6g.

- Present data at the annual NASA Human Research Program Investigators' Workshop and other scientific meetings.

- Prepare manuscripts for publication to peer reviewed journals.

The small wireless sensors developed in this study have been useful in other research efforts. Attached to the ankle, the sensor is currently being used in a study examining gait characteristics of female runners engaged in regular, long-distance training programs who are prone to injury. The goal is to reduce the rate of loading during running. A program has been developed to provide real-time feedback from the sensors as displayed on a screen in front of the runner. The runner can adjust their gait to a desirable data point in real-time, or take a portable data logger with them into the field and analyze the data post-run to see how the in-lab training has helped to re-train their running gait. Additional work has been proposed to use the sensors to assess the trade-off between hardware complexity and information density in comparing activity data from patients with bilateral osteoarthritis of the knee and unilateral trans-tibial amputations. We believe there are many other applications as well.

2) DEPLOY SOFTWARE UPDATES & INCREASE DATA THROUGHPUT: A comprehensive analysis of the wireless communication protocol and software configuration was conducted. It was determined that the sample frequency of the lower body sensor could be reduced from 1024Hz to 512Hz to assist in data throughput without significantly compromising science end points. The data structure was updated to an aggregation rate of 16, which also enhanced data throughput. These software updates have resulted in more reliable communication and more consistent data logging.

3) CONTINUED DATA ANALYSIS: Data analysis has been an on-going effort over the last year. Analysis of the data continues to suggest that the partial gravity environments of the moon or Mars will not alone be osteoprotective. Exercise will remain a necessary countermeasure in these environments. On-going work in this study will utilize the enhanced Daily Load Stimulus theory to help answer this question. It is certain that running in reduced gravity will still benefit from use of a subject load device to keep additional loading on the long-axis of the body and increase impact during locomotor exercises. Enhancements have been made to our activity recognition neural network programs. This will allow real-time reporting of a subject's daily activity and progress toward individual daily load stimulus target goals. Data sharing agreements have been made with members of NASA Glenn's Integrated Medical Model (IMM) team who is exploring input of the jump down data into their bone fracture risk model, and with Joern Rittweger of the German Aerospace Association to explore the energy losses experienced during static, stiff-legged hopping in 1g, 3/8g, and 1/6g environments. We will follow their progress in analysis of these data over the next year.

4) FLIGHT READINESS: Initial feasibility tests of the system aboard a parabolic flight have been performed courtesy of a colleague conducting a separate experiment. The system performed well during the flight. Plans to conduct a designated study aboard a dedicated parabolic research flight are being explored. Validation of the sensors in parabolic flight would advance our goals toward flight readiness. The sensors have successfully been interfaced with the Smartphone platform which can be used as a portable data logging system. Currently one sensor communicates with the phone at a time, and the feasibility of connecting two Bluetooth devices to one phone is being explored.

The original Project Objectives were to, namely:

1) Allow quantification of a crew health risk and.

2) Develop technologies to monitor a health risk.

The original Specific Aims of the project include:

1) To develop hardware based on Micro Electro Mechanical Systems technology (MEMS) that can unobtrusively monitor the accelerations applied to the body and interface with an ambulatory monitor.

2) To extend the Daily Load Stimulus Algorithm to account for recent developments in bone mechanobiology, to incorporate accelerometric signals, and to write software to perform this analysis in real-time.

3) To demonstrate the feasibility and validity of the approach in 1g, in 1/6g in the eZLS, and in the 1/6g lunar bed rest analog.

4) To integrate the hardware and software into a package suitable for flight development.

To date, the following goals and objectives have been met:

The wireless Daily Load Sensor has been designed, manufactured, and tested in a series of studies. The Daily Load Stimulus Algorithm has been extended to account for recent developments in bone mechanobiology, particularly the issue of stimulus saturation. The sensor system has been tested in 1g, 1/6g, and 3/8g environments. NASA HRP has decided to not utilize the lunar bed rest model, and as such the system has not been tested in that analog. A large data collection period has been completed at the end of Project Year 2, and initial data analysis is starting to reveal potential crew health risks to bone health maintenance in reduced gravity environments.

After validation in the enhanced Zero Gravity Locomotion Simulator (eZLS) at NASA Glenn Research Center, a deliverable of this project will be a system, the Daily Load Sensor, including a foot-mounted and a waist-mounted unit that will wirelessly transmit signals to a portable data logger that could potentially be used to collect data on other physiological systems simultaneously. Onboard software with visual feedback will determine how much additional exercise is required each day to maintain bone homeostasis. This high Technology Readiness Level (TRL=7) project combines theory, experimentation and hardware development to produce a device that will be a critical component in the effort to maintain bone health during lunar missions. The project is a collaborative effort between the University of Washington, the Exercise Countermeasures Laboratory at NASA Glenn Research Center, ZIN Technologies and the University of California, San Francisco.

Key findings to date include:

• Foot forces suggest IVA lunar and Martian locomotion (walking, running, loping, hopping) alone may not be osteoprotective.

• Foot forces suggest that a simulated lunar EVA (body weight plus ~200 Earth lbs of suit mass) locomotion may not be osteoprotective.

• Foot forces suggest that a simulated Martian EVA (body weight plus ~200 Earth lbs of suit mass) locomotion may provide adequate loading under some locomotion conditions to be osteoprotective depending on the duration of the activities.

• Lunar and Martian hopping and loping result in higher foot forces than walking. Running provides the highest foot forces in any one environment as compared to other locomotor activities.

• The eZLS provides simulation of reduced gravity environments and capability to utilize gravity replacement loads, similar to the treadmills in use on ISS.

• An Artificial Neural Network can precisely recognize lunar locomotor activity, which is useful during remote monitoring scenarios.

During year three of the project, we plan to achieve the following:

• Enhance the Activity Recognition library to include automated detection of additional mission critical tasks such as ladder climb, rock translation, platform jump down, squat exercise, and obstacle avoidance.

• Test the interface between our wireless sensors and the BioNet software framework in the laboratory setting, for future automated data management aboard the International Space Station (ISS).

• Determine feasibility of flight integration with hardware in current configuration, and asses need for modification to allow utilization aboard the ISS.

• Develop an ambulatory data logging system that is capable of communicating with the wireless sensors and the BioNet software framework.

• Instrument subjects who are participating in an on-going bed rest campaign conducted by our laboratory examining treadmill exercise countermeasures against bone demineralization.

At the end of year two we have made significant progress on each item, as described below.

1) HARDWARE DESIGN REVIEW AND MAINTENANCE: The Hardware Design Review was held at the beginning of Year 2. At that time, it was decided that the foot sensor would be encapsulated in a hard epoxy resin. The sampling rates of the waist and foot sensors were set to 256HZ and 1024Hz, respectively. ZIN Technologies continued to support hardware maintenance over Year 2 of the project, providing additional sensors upon request and re-calibration of the sensors half-way through the study. A software update was also performed to increase the resolution of the timestamp. Sensor battery chargers were replaced periodically throughout the study.

2) DATA COLLECTION & ANALYSIS: A total of 38 subjects were screened and consented for this study. A total of 25 subjects were enrolled in the study. Subjects were asked to perform a variety of mission critical tasks while positioned in the eZLS. These activities were performed in either shoes or boots, and at 1g, 1/6g, or 3/8g simulated gravity environments and included walking (1MPH), running (6MPH), loping (2MPH), hopping (2MPH), ladder climb, platform jump-down, 40lb rock carry, obstacle avoidance, and squat exercise. Data collection was completed in August 2010. Initial results were presented at the American Society of Biomechanics Annual Meeting and focused on locomotor activities (walking, running, loping, hopping).

3) EZLS FACILITY MAINTENANCE: NASA Glenn Research Center houses the enhanced Zero-gravity Locomotion System (eZLS) on which our hardware was tested. Specialty hardware needs were identified and engineers were tasked with the fabrication and integration of the ladder step, jump-down platform, obstacle avoidance hardware, and 40lb rock load. A customized passive pneumatic Subject Load Device (PP-SLD) was installed for Phase II of this study (See attached report for detailed description; Appendix B). The force plate on the instrumented treadmill was conditioned for low-gravity load usage, and the controller software was updated to include a reset zero function (See attached report for detailed description; Appendix B). All eZLS instrumentation was calibrated before the study began.

4) IRB MODIFICATIONS: Following full board approval from both the University of Washington IRB and the NASA CPHS, a number of modifications were requested and approved in Year 2 of the project, which included:

1) Increased subject recruitment numbers from 30 to 47.

2) Increased data sharing capability.

3) Record subject leg length for calculation of the Froude Number.

4) Decreased walking speed from 3MPH to 1MPH.

5) Included the 1g squat exercises in Phase II.

At the end of year one we have made significant progress on each item, as described below. 1) Hardware specification and development & 2) prototype fabrication: Science requirements have been identified and have aided in sensor identification. ZIN Technologies, a subcontractor on this grant, has designed a breadboard unit utilizing components which minimize size, mass, and power requirements while maximizing battery life and data transfer capabilities. Breadboard design and physical layout of components is complete. PCB assembly is scheduled to be complete before the end of Year 1. Completion of the prototype unit will allow for initial data collection and review of the products ability to meet scientific objectives. The hardware Preliminary Design Review is scheduled to occur at the beginning of Year 2.

3) Facility readiness (including fabrication of support hardware to meet scientific endpoints): NASA Glenn Research Center houses the enhanced Zero-gravity Locomotion System (eZLS) on which our prototype hardware will be tested in a series of two pilot studies. The third specific aim of this project is "To demonstrate the feasibility and validity of the approach in 1g, in 1/6g in the eZLS, and in the 1/6g lunar bedrest analog". Utilization of the eZLS provides the means to test the equipment in simulated microgravity and 1/6g environments. Another goal of the project is to test the hardware while typical lunar activities are being performed. A list of lunar activities has been adapted from NASA's Functional Task Test protocol for inclusion in our study. These include a rock translation, platform jump down test, and obstacle course maneuvering. Specialty hardware needs have been identified and engineers have been assigned to the task of fabrication and integration. Funds are allocated in the Year 2 for fabrication and integration of these lunar task items.

4) IRB approval: Full applications for human subject approval were sent to the University of Washington and the NASA CPHS Intstitutional Review Boards (IRBs) for approval. To date, full approval from the University of Washington has been granted and conditional approval has been granted by CPHS with minor stipulations. Full approval from CPHS is anticipated before the start of Year 2.