Intelligent Millimeter-Wave System for Human Activity Monitoring for Telemedicine.

Telemedicine has the potential to improve access and delivery of healthcare to diverse and aging populations. Recent advances in technology allow for remote monitoring of physiological measures such as heart rate, oxygen saturation, blood glucose, and blood pressure. However, the ability to accurately detect falls and monitor physical activity remotely without invading privacy or remembering to wear a costly device remains an ongoing concern. Our proposed system utilizes a millimeter-wave (mmwave) radar sensor (IWR6843ISK-ODS) connected to an NVIDIA Jetson Nano board for continuous monitoring of human activity. We developed a PointNet neural network for real-time human activity monitoring that can provide activity data reports, tracking maps, and fall alerts. Using radar helps to safeguard patients' privacy by abstaining from recording camera images. We evaluated our system for real-time operation and achieved an inference accuracy of 99.5% when recognizing five types of activities: standing, walking, sitting, lying, and falling. Our system would facilitate the ability to detect falls and monitor physical activity in home and institutional settings to improve telemedicine by providing objective data for more timely and targeted interventions. This work demonstrates the potential of artificial intelligence algorithms and mmwave sensors for HAR.

Towards a Low-Cost Solution for Gait Analysis Using Millimeter Wave Sensor and Machine Learning.

Human Activity Recognition (HAR) that includes gait analysis may be useful for various rehabilitation and telemonitoring applications. Current gait analysis methods, such as wearables or cameras, have privacy and operational constraints, especially when used with older adults. Millimeter-Wave (MMW) radar is a promising solution for gait applications because of its low-cost, better privacy, and resilience to ambient light and climate conditions. This paper presents a novel human gait analysis method that combines the micro-Doppler spectrogram and skeletal pose estimation using MMW radar for HAR. In our approach, we used the Texas Instruments IWR6843ISK-ODS MMW radar to obtain the micro-Doppler spectrogram and point clouds for 19 human joints. We developed a multilayer Convolutional Neural Network (CNN) to recognize and classify five different gait patterns with an accuracy of 95.7 to 98.8% using MMW radar data. During training of the CNN algorithm, we used the extracted 3D coordinates of 25 joints using the Kinect V2 sensor and compared them with the point clouds data to improve the estimation. Finally, we performed a real-time simulation to observe the point cloud behavior for different activities and validated our system against the ground truth values. The proposed method demonstrates the ability to distinguish between different human activities to obtain clinically relevant gait information.

Impact of an ankle foot orthosis on reactive stepping in young adults.

BACKGROUND: Ankle-foot orthoses (AFOs) have been shown to improve gait and static balance in individuals with lower extremity weakness and instability. However, the effects of AFOs on dynamic balance reactions including reactive stepping responses are not well known. Therefore, the purpose of this study was to determine the effects of an AFO on reactive stepping responses in healthy young adults. RESEARCH QUESTION: Does an AFO alter reactive stepping responses in healthy young adults? METHODS: Twenty healthy young adults completed 10 reactive stepping trials using a lean-and-release system for each of three AFO conditions: 1) no AFO, 2) AFO on left leg and 3) AFO on right leg. Trials were recorded using 3D motion capture and force plates. Stepping limb preference and temporal, spatial, and kinematic variables were measured. Differences between conditions were determined by a one-way ANOVA with a Tukey post-hoc. RESULTS: With no AFO, participants demonstrated a preference for stepping with the right leg, 7.0 ± 3.9 of 10 trials. With an AFO on the right leg, this preference decreased to 5.7 ± 4.4 (p = 0.03). With an AFO on the left leg, this preference increased to 8.1 ± 3.3 (p = 0.03). Reaction times were not significantly different between conditions, but participants took a significantly shorter reactive step with the leg wearing the AFO. Peak ankle, knee, and hip joint angles were significantly less with the AFO on the stepping limb compared to the stance limb. SIGNIFICANCE: This study shows that AFO use can influence reactive stepping limb preference and stepping limb kinematics in healthy young adults. These results can inform future research on AFO users with gait impairments. These finding may also be helpful in developing interventions to address the specific effects of an AFO on reactive stepping responses.

The effects of a positional feedback device on rollator walker use: a validation study.

BACKGROUND: According to clinical guidelines, rolling walker users should walk with their feet between the posterior wheels of the walker; however approximately 50% of users do not. OBJECTIVE: To describe the development and effects of a custom device designed to attach to a walker and provide visual feedback to encourage improved user position. METHODS: Fourteen older adults participated in this study to validate the effects of this device when a 10% decrease in the users' habitual distance away from the walker was encouraged via feedback. Users' relative distances were recorded using a non-contact distance sensor within the device, while kinematics were measured using commercial wearable wireless inertial sensors. RESULTS: Individuals were able to ambulate on average 20% closer or more to their walker when prescribed the visual feedback. This was primarily achieved through a reduction in shoulder flexion. Trunk and cervical postures were less generalizable as only small and variable changes were observed. CONCLUSIONS: These findings suggest that the device has promise, as individuals attended to the device and walked in a position closer to that recommended by clinical guidelines. The device did not appear to improve posture. Future work is needed to determine long-term effects.

Age Stratification and Sample Entropy Analysis Enhance the Limits of Stability Tests for Older Adults.

The objective of this study was to determine if a foam testing condition and/or nonlinear analysis methods can be used to identify differences between age stratified subgroups of older adults when conducting the Limits of Stability assessment. Ninety older adults participated in this study. A force plate was used to record center of pressure data during Limits of Stability testing on a firm and foam surface. Participants were grouped into age-stratified subgroups: young-old (60-69 years), middle-old (70-79 years), and old-old (80+ years). Anterior-posterior (A/P) and medial-lateral (M/L) sway ranges and sample entropy values were calculated. The young-old group had significantly larger A/P and M/L sway ranges than the old-old group on both surfaces. A/P sample entropy increased significantly with age. M/L sample entropy increased significantly with age between the young-old and old-old and the middle-old and old-old groups. Sample entropy values between surfaces significantly differed for all groups. These results indicate Limits of Stability differences occur between older adults of different age groups and should be taken into consideration for clinical and research testing. Nonlinear analysis may help further identify differences in Limits of Stability performance while use of a foam surface is of limited additional value.

Identification of key outcome measures when using the instrumented timed up and go and/or posturography for fall screening.

The Timed Up and Go (TUG) has been commonly used for fall risk assessment. The instrumented Timed Up and Go (iTUG) adds wearable sensors to capture sub-movements and may be more sensitive. Posturography assessments have also been used for determining fall risk. This study used stepwise logistic regression models to identify key outcome measures for the iTUG and posturography protocols. The effectiveness of the models containing these measures in differentiating fallers from non-fallers were then compared for each: iTUG total time duration only, iTUG, posturography, and combined iTUG and posturography assessments. One hundred and fifty older adults participated in this study. The iTUG measures were calculated utilizing APDM Inc.'s Mobility Lab software. Traditional and non-linear posturography measures were calculated from center of pressure during quiet-standing. The key outcome measures incorporated in the iTUG assessment model (sit-to-stand lean angle and height) resulted in a model sensitivity of 48.1% and max re-scaled R2 value of 0.19. This was a higher sensitivity, indicating better differentiation, compared to the model only including total time duration (outcome of the traditional TUG), which had a sensitivity of 18.2%. When the key outcome measures of the iTUG and the posturography assessments were combined into a single model, the sensitivity was approximately the same as the iTUG model alone. Overall the findings of this study support that the iTUG demonstrates greater sensitivity than the total time duration, but that carrying out both iTUG and posturography does not greatly improve sensitivity when used as a fall risk screening tool.

Manual and Cognitive Dual Tasks Contribute to Fall-Risk Differentiation in Posturography Measures.

Falls occur in 33% of older adults each year, some leading to moderate to severe injuries. To reduce falls and fall-related injuries, it is important to identify individuals with subtle risk factors elevating their likelihood of falling. The objective of this study was to determine how postural sway measures differed between fallers and nonfallers under standard and dual-task conditions. Quietstanding posturography measures were collected from 150 older adults during standard, cognitive, manual, and cognitive+manual tasks, and analyzed through traditional and nonlinear analyses. Of the traditional measures, M/L sway range and 95% confidence ellipse sway area showed statistically significant differences in all 4 test conditions between fallers and nonfallers. Although the manual dual task showed the most stable balance, effect sizes demonstrated larger differences between fallers and nonfallers. Nonlinear analysis revealed M/L sample entropy and M/L α-scaling exponent differentiating between fallers and nonfallers, with the cognitive task demonstrating larger differences. Based on the results, it is recommended to: (1) apply M/L sway range and 95% confidence ellipse area, (2) use the manual task to differentiate between fallers and nonfallers when using traditional analyses, and (3) use the cognitive task and M/L alpha and M/L sample entropy when using nonlinear analyses.

The Effects of Backwards Running Training on Forward Running Economy in Trained Males.

Backwards running (BR) results in greater cardiopulmonary response and muscle activity compared with forward running (FR). BR has traditionally been used in rehabilitation for disorders such as stroke and lower leg extremity injuries, as well as in short bursts during various athletic events. The aim of this study was to measure the effects of sustained backwards running training on forward running economy in trained male athletes. Eight highly trained, male runners (26.13 ± 6.11 years, 174.7 ± 6.4 cm, 68.4 ± 9.24 kg, 8.61 ± 3.21% body fat, 71.40 ± 7.31 ml·kg(-1)·min(-1)) trained with BR while harnessed on a treadmill at 161 m·min(-1) for 5 weeks following a 5-week BR run-in period at a lower speed (134 m·min(-1)). Subjects were tested at baseline, postfamiliarized, and post-BR training for body composition, a ramped VO2max test, and an economy test designed for trained male runners. Subjects improved forward running economy by 2.54% (1.19 ± 1.26 ml·kg(-1)·min(-1), p = 0.032) at 215 m·min(-1). VO2max, body mass, lean mass, fat mass, and % body fat did not change (p > 0.05). Five weeks of BR training improved FR economy in healthy, trained male runners without altering VO2max or body composition. The improvements observed in this study could be a beneficial form of training to an already economical population to improve running economy.

Using Posturography to Examine the Immediate Effects of Vestibular Therapy for Children with Autism Spectrum Disorders: A Feasibility Study.

AIMS: The primary objective of this study was to determine the feasibility of using posturography to monitor acute changes in postural control induced by a Sensory Integration (SI) therapy intervention. A secondary objective was to identify which posturography outcome parameters, tests conditions and data analysis methods might be most useful in identifying post-intervention changes. METHODS: Five children with Autism Spectrum Disorder (ASD) and five children with typical development (TD) participated in a 10 min vestibular swing activity and had their postural stability evaluated pre- and post-intervention under four different sensory testing conditions. Sway ranges, mean sway velocity, sway root mean square (RMS), and sample entropy were calculated from center of pressure (COP) data. RESULTS: All five children with ASD demonstrated decreased mean sway velocity in the eyes open/flat plate condition post-intervention with an average decrease of 5.87 ± 2.69 mm/s. Four of the five children with ASD demonstrated an increase in RMS and a decrease in anterior/posterior sample entropy post-intervention in the eyes closed, foam pad condition and eyes open, flat plate condition respectively. CONCLUSION: Posturography may be useful for assessing acute physiologic responses to an SI therapy intervention and warrants further investigation.

Effects of heat stress and sex on pacing in marathon runners.

Recent research suggests that women tend to exhibit less of a precipitous decline in run velocity during the latter stages of a marathon than men when the covariates of age and run time are controlled for. The purpose of this study was to examine this sex effect with the added covariate of heat stress on pacing, defined as the mean velocity of the last 12.2 km divided by the mean velocity of the first 30 km. A secondary purpose of this investigation was to compare the pacing profiles of the elite men and women runners and the pacing profiles of the elite and nonelite runners. Subjects included 22,990 men and 13,233 women runners from the 2007 and 2009 Chicago marathons for which the mean ambient temperatures were 26.67° C and 2.77° C, respectively. Each 5-km split time was measured via an electronic chip worn on the participants' shoe. Multiple regression analysis indicated that age, sex, heat stress, and overall finish time (p < 0.01 for each) were simultaneous independent elements of pacing. Nonelite women were consistently better pacers than nonelite men in both marathons, and this sex difference was magnified from cold to warm race temperatures. No difference (p < 0.05) in pacing was found between elite men and women runners. Elite men and women had enhanced pacing over their nonelite counterparts. In hotter temperatures, coaches of novice runners should advise their athletes to implement a slower initial velocity to maintain or increase running velocity later in the race.

Measures of balance performance are affected by a rested versus fatigued testing condition in people with multiple sclerosis.

OBJECTIVE: To investigate the effects of a rested and fatigued testing condition on measures of balance in people with multiple sclerosis (MS). DESIGN: Prospective observational study. SETTING: Academic research laboratory. PARTICIPANTS: Fifteen ambulatory adults with relapsing remitting or secondary progressive MS with moderate disability. METHODS: Measures of postural control, dynamic balance, and fatigue were assessed during a "rested" and "fatigued" testing condition on separate days. MAIN OUTCOME MEASUREMENTS: Static posturography was used to assess postural control. The Mini-Balance Evaluation Systems Test and Dynamic Gait Index were used to assess dynamic balance. The chronic fatigue level was measured by using the Fatigue Severity Scale, and acute fatigue was assessed by using a visual analog scale at multiple time points during testing. RESULTS: During the fatigued condition, the participants reported a significant (P < .05) increase in acute fatigue levels (visual analog scale) and demonstrated a significant (P < .05) decrease in performance on measures of postural control and dynamic balance when compared with the rested condition. However, no significant relationship was identified between the changes in fatigue and changes in balance performance. CONCLUSIONS: Measures of balance performance and acute fatigue may be affected by rested versus fatigued testing conditions in people with MS. However, no significant relationships between changes in balance and fatigue were identified. Therefore, clinicians should use caution when making assumptions about the influence of fatigue on balance performance and should test for individual responses. Clinicians and researchers should also consider controlling for factors such as time of day and prior physical activity when administering standardized measures of balance in people with MS.

A group kickboxing program for balance, mobility, and quality of life in individuals with multiple sclerosis: a pilot study.

BACKGROUND AND PURPOSE: Balance and mobility impairments are common in individuals with multiple sclerosis (MS). The primary purpose of this pilot study was to evaluate the feasibility of a 5-week group kickboxing program and to measure changes in balance, mobility, and quality of life in individuals with MS associated with this training. METHODS: This single-group repeated-measures study involved a convenience sample of 15 individuals with MS who had minimal to moderate levels of disability and were recruited from the community. Eleven participants completed all phases of testing and training. The intervention was a 5-week group kickboxing program performed 3 times per week. Outcome measures were assessed 5 weeks prior to the intervention, 1 week prior to the intervention, and within 1 week of completing the intervention. Outcome measures include gait speed, Timed Up & Go test, Berg Balance Scale, Dynamic Gait Index, Mini-BESTest, Activities Specific Balance Confidence scale, and the Multiple Sclerosis Quality of Life Survey. RESULTS: There were significant improvements in gait speed, some clinical measures of balance, and balance confidence following the intervention but no changes were observed in health-related quality of life. There were no unanticipated adverse events and compliance was high. CONCLUSION: Group kickboxing appears to be a feasible exercise activity for individuals with MS that may lead to improvement in select measures of balance and mobility. However, the clinical relevance of these findings is yet to be determined. Further investigation of this novel intervention may be warranted.

Feasibility and effects of a group kickboxing program for individuals with multiple sclerosis: a pilot report.

BACKGROUND AND PURPOSE: Balance and mobility impairments are common in persons with multiple sclerosis (MS). The primary purpose of this pilot program was to evaluate the feasibility and the effects of group kickboxing on balance and mobility in individuals with MS. METHODS: Four individuals with relapsing-remitting or secondary progressive MS participated in a group kickboxing program two times per week for 8 weeks. Outcome measures included the Berg Balance Scale (BBS), Dynamic Gait Index (DGI), Timed Up and Go (TUG), walking speed and the Activities Specific Balance Confidence Scale (ABC). RESULTS: Following training, 3 of 4 participants had improvements in BBS performance. All participants demonstrated improvements in the DGI. Changes in the TUG, ABC, and walking speed were more variable. CONCLUSION: Group kickboxing appears to be a feasible exercise activity for individuals with MS and may lead to improvement in select measures of balance. Further investigation may be warranted.

Systematically controlling for the influence of age, sex, hertz and time post-whole-body vibration exposure on four measures of physical performance in community-dwelling older adults: a randomized cross-over study.

Though popular, there is little agreement on what whole-body vibration (WBV) parameters will optimize performance. This study aimed to clarify the effects of age, sex, hertz and time on four physical function indicators in community-dwelling older adults (N = 32). Participants were exposed to 2 min WBV per session at either 2 Hz or 26 Hz and outcome measures were recorded at 2, 20 and 40 min post-WBV. Timed get up-and-go and chair sit-and-reach performances improved post-WBV for both sexes, were significantly different between 2 Hz and 26 Hz treatments (P ≤ 0.05) and showed statistically significant interactions between age and gender (P ≤ 0.01). Counter movement jump and timed one-legged stance performances showed a similar but non-significant response to 2 Hz and 26 Hz treatments, though male subjects showed a distinct trended response. Age and gender should be statistically controlled and both 2 Hz and 26 Hz exert a treatment effect.

Use of an elliptical machine for improving functional walking capacity in individuals with chronic stroke: a case series.

BACKGROUND AND PURPOSE: Decreased functional walking capacity is a common consequence of stroke. Identifying practical and cost-effective methods to improve walking in individuals with stroke is an important goal of rehabilitation professionals. PARTICIPANTS: Participants were 3 men with chronic (>6 month) stroke, who could walk on level surfaces either without an assistive device or with a single-point cane. INTERVENTION: Participants trained 2 to 3 times per week for 8 weeks, using an elliptical machine. The training target was 20 minutes of uninterrupted training, while maintaining predetermined parameters of heart rate and perceived exertion. OUTCOMES: Outcome measures assessed before and after training included habitual and fast gait speed, 6-minute walk test (6MWT), Timed "Up & Go" test, and Berg Balance Scale. Following training there was no change in walking speed. There was no change in 6MWT performance for participants 1 and 2. While participant 3 showed a 25% improvement in 6MWT, this change did not meet the minimal detectable change for walking speed in individuals with stroke. All participants demonstrated improved Berg Balance Scale performance (9%-28%), with participant 1 exceeding the minimal detectable change in this measure. Timed Up & Go test performance improved by 5% to 15% in all participants. DISCUSSION: Elliptical training appears to be a safe and feasible training alternative for ambulatory individuals with chronic stroke. Training 2 to 3 days per week resulted in no improvements in walking speed; however, participants did demonstrate variable improvements in endurance, balance, and functional mobility. It is possible that a higher training frequency and/or training speed are required to influence walking performance in individuals who are ambulatory. Equipment design, principles of exercise prescription, and participant characteristics should be considered when selecting elliptical training as an intervention.

The effects of whole-body vibration training in aging adults: a systematic review.

BACKGROUND AND PURPOSE: Whole-body vibration (WBV), has become increasingly popular as a form of exercise training. WBV involves the application of a vibratory stimulus to the entire body as opposed to local stimulation of specific muscle groups. The purpose of this review was to assess the evidence concerning the effectiveness of WBV training studies on bone density, muscle performance, balance, and functional mobility in older adults and to discuss potential precautions, safety concerns, and practical clinical considerations of WBV. METHODS: A literature search of online databases was conducted and methodological quality assessment was performed using the critical appraisal scales developed by Sackett and Jadad on the WBV articles that met the predetermined inclusion criteria. RESULTS: The initial search resulted in the retrieval of 196 potential articles. One additional article was found by manual search. After review, 13 studies were identified that met the predetermined selection criteria. DISCUSSION: Much of the WBV research to date is methodologically weak and should be interpreted with caution. Study protocols have used widely variable WBV parameters which also complicates the studies' interpretation. Some but not all of the studies in this review reported similar improvements in muscle performance, balance, and functional mobility with WBV as compared to traditional exercise programs. Bone studies consistently showed that WBV improved bone density in the hip and tibia but not in the lumbar spine. CONCLUSION: Additional studies are needed to determine safe and effective parameters for WBV training in older adults.

Acute effects of whole-body vibration on lower extremity muscle performance in persons with multiple sclerosis.

BACKGROUND AND PURPOSE: Whole-body vibration (WBV) is a relatively new form of exercise training that may influence muscle performance. This study investigated the acute effects of high- (26 Hz) and low- (2 Hz) frequency WBV on isometric muscle torque of the quadriceps and hamstrings in persons with multiple sclerosis. PARTICIPANTS AND METHOD: Fifteen individuals (mean age = 54.6 years, SD = 9.6) with multiple sclerosis and Expanded Disability Status Scale scores ranging from 0 to 6.5 (mean = 4.2, SD = 2.3) participated in this randomized, crossover study. After baseline measures of isometric quadriceps and hamstring muscle torque, subjects were exposed to 30 seconds of WBV at either 2 or 26 Hz. Torque values were measured again at one, 10, and 20 minutes after vibration. Subjects returned one week later to repeat the same protocol at the alternate vibration frequency. RESULTS: There were no significant differences in isometric torque production between the 2- and 26-Hz WBV conditions. There was also no significant difference between baseline torque values and those measured at one, 10, and 20 minutes after either vibration exposure. However, there was a consistent trend of higher torque values after the 26-Hz WBV when compared with the 2-Hz condition for both quadriceps and hamstring muscles. DISCUSSION AND CONCLUSION: Although not statistically significant, peak torque values for both quadriceps and hamstring muscles were consistently higher after 30 seconds of WBV at 26 vs 2 Hz. Whether WBV presents a viable treatment option as either a warm-up activity or a long-term exercise intervention is yet to be determined. Future studies should include a wider variety of WBV parameters and the use of functional outcome measures.