The second objective was accomplished by developing the VETSS, using the Microsoft Kinect depth-based camera system. Basic software architecture was developed before collecting deadlift exercise input data on 20 participants. Integrating feedback from experienced strength and conditioning coach, a machine learning algorithm was used to pre-train the VETSS to identify five common deadlift errors. Subsequently, 16 new participants completed four sessions separated by one to two weeks. The first session was used to instruct subjects in proper deadlift technique; this was followed a week later by a one-rep maximum (1RM) determination and collection of a well-performed reference deadlift with a 3D motion capture system. Participants were randomized to either a control group (CON) with no feedback or a group that utilized VETSS (EXP).
The third objective of this project was to test the integration of an existing VES tool for its effectiveness to improve performance in a single bout of high intensity treadmill exercise, and adherence and performance during a one-month metabolic conditioning program. Two randomized controlled trials were completed to accomplish this objective. First, a randomized cross-over study design was employed for 23 participants. After determining their peak oxygen consumption (VO2peak), subjects completed three timed run-to-failure (TRF) tests at the same high intensity. TRF tests were conducted with the following conditions in random order: 1) exercise with no external stimuli (CON), 2) exercise with subject-selected music (MUS), or 3) exercise with subject-selected music and VES. During the TRF, test time(s), rating of perceived exertion (RPE), heart rate (HR), oxygen consumption (VO2), and respiratory exchange ratio (RER) were collected. The VES condition allowed users to select from 18 pre-recorded environments from scenic locations around the world and was displayed on a 121 cm screen in front of the participant.
NextGen participants who were physically active but not runners (N=18) completed the second phase of the project. They were matched for VO2 and body mass index (BMI) and randomly assigned to a self-selected music (MUS) or with music and VES group. Subjects completed a week of testing that consisted of TRF trials. Subsequently, they completed three or five days of treadmill exercise per week for four weeks (12-20 sessions). Workout durations ranged from 10-30 minutes.
Results indicate that participants took 21% less time to complete exercise sessions in the NextGen user interface (UI) condition than in the ARED UI condition. Additionally, measures of learnability show that the design of the NextGen UI better assisted participants in learning the system, evidenced by a 30% decrease in task completion time between trials for the NextGen UI, whereas the ARED UI showed at 25% decrease in trial completion time.
From the VETSS data investigators were able to conclude that exercise using the VETSS may improve hip kinematics and reduce spinal flexion; however, it does not seem to improve knee flexion kinematics. Deadlift exercise with the VETSS may provide corrective feedback, helping to prevent more severe spinal flexion and improving the consistency of hip kinematics.
In addition, the results indicate that participants using VES ran for significantly longer than CON, but MUS was not significantly different than VES or CON. HR was elevated in VES and MUS compared to CON. No other metabolic or perceptual differences were evident. Participants in both groups completed a similar number of intervals and the average peak HR was similar between groups in training. Peak RER during VO2peak tests were similar between groups at baseline and did not change during the post test. VO2peak increased in VES post-training but there was no change in MUS. Peak ventilation also increased in both VES and MUS but there was no between group difference. HR during the post-test significantly decreased in VES.
|Mission||Launch/Start Date||Landing/End Date||Duration|