Estimating pedestrian walking speed at street crossings using the YOLOv4 and deep SORT algorithms: Proof of principle.

There is evidence that existing standards for signal timing do not provide enough time for many pedestrians to safely cross intersections. Yet, current methods for studying this problem rely on inefficient manual observations. The objective of this work was to determine if the YOLOv4 and Deep SORT computer vision algorithms have the potential to be incorporated into automated measurement systems to measure and compare pedestrian walking speeds at one-stage and two-stage street crossings captured in birds-eye-view video. Walking speed was estimated for 1018 pedestrians at single-stage (591 pedestrians) and two-stage (427 pedestrians) street crossings. Pedestrians in the one-stage crossing were found to be significantly slower than pedestrians who crossed the two-stage crossing in one signal (1.19 ± 0.50 vs. 1.31 ± 0.49 m/s, p < 0.001). This proof of principle study demonstrated that the YOLOv4 and Deep SORT approaches are promising for estimating pedestrian walking speed.

Investigating the Effect of Addition of Cervical Proprioceptive Training to Conventional Physiotherapy on Visual, Vestibular and Proprioceptive Dependency of Postural Control in Patients with Chronic Non-Specific Neck Pain: A Randomized Controlled Clinical Trial.

Objectives: While cervical proprioception deficit has been suggested as a contributing factor to clinical consequences of chronic non-specific neck pain (CNSNP), the effect of addressing such impairments on postural control strategies has remained unexplored. The aim of this study was to compare the response of the postural control system to alteration of sensory afferents in CNSNP with asymptomatic individuals. Furthermore, we examined whether proprioceptive training would yield superior outcomes to routine physiotherapy for improvement of postural control, pain and disability. Methods: Center of pressure (CoP) variables of sixty CNSNP patients equally distributed in any of the proprioception-specific or conventional physiotherapy groups and 30 asymptomatic participants were evaluated under four standing conditions:1) normal, 2) foam, 3) cervical extension/eyes open and 4) cervical extension/eyes closed standing. Results: CoP anteroposterior range and anteroposterior and mediolateral velocity in patients were significantly higher than the control group under condition 2 (P<0.05). Patients also demonstrated lower anteroposterior lyapunov exponent under conditions 2 and 4 (P<0.05). Both interventions significantly decreased anteroposterior range and anteroposterior velocity(P<0.05). Anteroposterior lyapunov exponent also increased under condition 2 (P<0.05).. After the interventions, CoP anteroposterior range and anteroposterior velocity were significantly lower in the proprioceptive exercise group than the conventional physiotherapy group (P<0.05). Anteroposterior lyapunov exponent was also significantly higher in the proprioceptive exercise group (P<0.05).This while there was no significant difference between these patients and control group participants in any of the CoP variables after intervention. Conclusion: Our results rejected the hypothesis that impaired neck proprioception in the presence of CNSNP is compensated by overweighting other sources of sensory afferent information. The findings also revealed that while proprioceptive exercises successfully returned postural strategies of CNSNP patients to those in asymptomatic participants, they do not add to clinical recovery of these patients.

Comparison of Postural Control of Females with and without Urinary Incontinence: A Case-Control Study.

Background: So far, there is much less information about the effects of urinary incontinence on postural control. Therefore the aim of this study is to investigate the differences in postural control using linear and non-linear analyses of the center of pressure (COP) time-series in anteroposterior (AP) and mediolateral (ML) directions between females with and without stress urinary incontinence (SUI). Methods: This case-control study included 22 continent females and 22 SUI females. In this study, static postural control during four different postural tasks was evaluated using a force plate. All participants performed separate 60-sec standing trials with eyes open in the empty bladder and full bladder conditions. Mean, range, velocity, area circle of COP displacements, and approximate entropy (ApEn) of COP time-series were calculated from the 60-sec standing trials for all participants. The independent sample t-test was also used to compare COP variables between the two groups and paired sample t-test was used to assess changes between the full bladder and empty bladder conditions within each group. The effect size of Cohen's d was used to assess the magnitude of the differences between the two groups. Results: The findings revealed a significant group × task interaction for the mean of ML displacement and ApEn of COP. SUI females showed more AP displacement range in the full bladder (pvalue= 0.020, effect size=0.74) and a higher velocity (empty bladder: p=0.040, effect size=0.63) (full bladder: p=0.020, effect size=0.75) than the continent group. Generally, the SUI females had lower ApEn than the continent females, although the differences were not significant. While the variables of COP were unaffected by bladder fullness in the continent group, the SUI group in full bladder condition experienced more AP range (p=0.030), and area circle (p=0.007) of COP sway in quiet standing. Conclusion: These results provide more support for the hypothesis that postural control can be impaired following SUI, although future investigations on this topic are recommended.

Effects of cognitive load on the amount and temporal structure of postural sway variability in stroke survivors.

This study aimed to investigate the variability in postural sway patterns during quiet standing in stroke survivors. The postural sway was measured in 19 stroke survivors, as well as 19 healthy demographically matched participants, at 3 levels of postural difficulty (rigid surface with closed and open eyes, and foam surface with closed eyes), and 3 levels of cognitive difficulty (without a cognitive task, easy and difficult cognitive tasks). Both linear analyses (the amount of postural sway variability, including the standard deviation of the COP velocity in both the anteroposterior (AP) and mediolateral (ML) directions), as well as non-linear analyses [the temporal structure of the COP variability, including % Recurrence, % Determinism, Shannon Entropy, Trend and the maximum diagonal line (D max)] were employed. The results revealed that the amount of variability of the postural sway of stroke survivors was significantly greater than that of healthy participants, along both the ML and AP directions, with the highest obtained during standing on foam with closed eyes. All measures of the temporal structure of the COP variability were significantly greater in stroke survivors, as compared to the control group, along the ML direction, but not along the AP direction. The cognitive error was significantly higher during difficult cognitive tasks, although it was neither affected by postural difficulty nor by group. The different results obtained for the amount and temporal structure of the COP variability in the AP and ML directions shed light on the intricate mechanisms employed by the CNS in post-stroke balance control, and suggest that effective rehabilitative and therapeutic strategies should be patient-specific, taking both the environment/surface as well as the specific protocols into consideration.

Investigating the effects of visual biofeedback therapy on recovery of postural balance in stroke patients using a complexity measure.

BACKGROUND AND PURPOSE: Postural balance deficit is one of the common post-stroke disabilities. Providing visual biofeedback while balance activities are performed is a way to improve postural balance disorders following stroke. But among the research publications, there is incoherency about the positive effects of visual biofeedback therapy. The purpose of this study was to investigate the effects of using visual biofeedback as an adjunct to physical therapy exercises on recovery of postural balance of stroke patients. MATERIALS AND METHODS: A total of 31 hemiplegic stroke patients were recruited in this study and randomly assigned into case and control groups. Both groups received conventional physical therapy interventions and balance training exercises. During balance training, the case group received visual biofeedback, whereas the control group did not receive visual information. Balance performance of stroke patients were examined quantitatively using the EquiTest testing system. Center of pressure data were collected before starting, during, and after completion of the rehabilitation program and a nonlinear complexity measure, approximate entropy (ApEn), calculated and used for the analysis. RESULTS: No significant between-group differences were detected after completion of the program. Noticeable increase was found in ApEn values of both groups along anterior-posterior direction, whereas no statistically significant improvement was found along mediolateral direction after rehabilitation. CONCLUSION: Both rehabilitation routines created advances in the postural control system of stroke patients. Visual biofeedback balance training did not produce extra advantage for balance ability of participants who received this treatment program in comparison with those who were treated without visual biofeedback.

The effects of a short-term memory task on postural control of stroke patients.

BACKGROUND: Many studies have been conducted on the changes in the balance capabilities of stroke patients. However, results regarding the effects of dual-task activities on postural control in these patients have been variable. OBJECTIVE: To evaluate the effects of a short-term memory task on the sway characteristics of stroke patients. METHOD: Center of pressure (COP) fluctuations were measured in three levels of postural difficulty (rigid surface with closed and open eyes and foam surface with closed eyes), as well as two levels of cognitive difficulty (easy and difficult). COP parameters included mean velocity, standard deviation of velocity in both medial-lateral (M.L) and anterior-posterior (A.P) directions, total phase plane portrait, area. Nineteen stroke patients and 19 gender, age, height, and weight matching healthy volunteers participated in this study. RESULTS: Our findings indicate that mean velocity (F = 14.21, P = 0.001), standard deviation of velocity in both M.L (F = 17.50, P = 0.000) and A.P (F = 11.03, P = 0.002) directions, total phase plane portrait (F = 44.12, P = 0.001), and area (F = 13.95, P = 0.01) of center of pressure of patients were statistically greater than normal subjects, while significant interaction of group × postural difficulty and postural × cognitive difficulty were observed for all parameters of postural sway. CONCLUSIONS: Different measures of postural sway showed complex response to postural and cognitive difficulties between stroke patients and normals. Cognitive error was not affected by the main effects of group and postural difficulty, while greatly increased at more difficult cognitive task (F = 75.73, P = 0.000).

Fundamental frequency changes of Persian speakers across the life span.

This study was designed to investigate changes in fundamental frequency (F0) across the life span in Persian speakers. Four hundred children and adults were asked to produce a sustained phonation of vowel /a/ and their voice samples were studied in 10 age groups. F0 was analyzed using the software Praat (Version 5.1.17.). The results revealed that (1) the mean F0 in both sexes decreases from childhood to adulthood; (2) significant F0 differences between boys and girls begin at the age of 12 years; and (3) the range of F0 changes in the life span is greater in men (178.38 Hz) than in women (113.57 Hz). These findings provide new data for Persian-speaking children, women, and men and could be beneficial for Iranian speech and language pathologists.

Spectral characteristics of postural sway in diabetic neuropathy patients participating in balance training.

BACKGROUND: The aim of the present study is to have a detailed frequency analysis about the effect of balance training with respect to reactive movement strategies and sensory strategies in type 2 diabetic neuropathy (DN) patients. Also understand changes in the role of each postural subsystem for controlling quiet standing after balance training. METHODS: A group of 19 patients were included in the quasi experimental, time- series study. Total frequency power, 99% power frequency, centroidal frequency and frequency spectrum in the intervals between 0.01-0.1, 0.1-0.5, 0.5-1 Hz and 1-3 Hz are reported. The training protocol consisted two patterns of limits of stability trainings, three approaches in weight shifting trainings and one stable standing practice on the biodex stability system. RESULTS: Repeated measure ANOVA analysis and the LSD test indicated significant differences for the eyes open ML- frequency power and ML-FFT sway power within low-medium (0.1-0.5 HZ) frequencies. CONCLUSIONS: Decrease in postural sway at low-medium frequencies showed lower reliance on vestibular system. Also, better controlling hip muscles after balance training relieve DN patients' requirement to more exploratory sway as a compensatory strategy and showed better balance performance after balance training in DN patients. TRIAL REGISTRATION: UMIN-CTR Search Clinical Trials: UMIN000004485.