Convolutional Neural Network Approaches in Median Nerve Morphological Assessment from Ultrasound Images.

Ultrasound imaging has been used to investigate compression of the median nerve in carpal tunnel syndrome patients. Ultrasound imaging and the extraction of median nerve parameters from ultrasound images are crucial and are usually performed manually by experts. The manual annotation of ultrasound images relies on experience, and intra- and interrater reliability may vary among studies. In this study, two types of convolutional neural networks (CNNs), U-Net and SegNet, were used to extract the median nerve morphology. To the best of our knowledge, the application of these methods to ultrasound imaging of the median nerve has not yet been investigated. Spearman's correlation and Bland-Altman analyses were performed to investigate the correlation and agreement between manual annotation and CNN estimation, namely, the cross-sectional area, circumference, and diameter of the median nerve. The results showed that the intersection over union (IoU) of U-Net (0.717) was greater than that of SegNet (0.625). A few images in SegNet had an IoU below 0.6, decreasing the average IoU. In both models, the IoU decreased when the median nerve was elongated longitudinally with a blurred outline. The Bland-Altman analysis revealed that, in general, both the U-Net- and SegNet-estimated measurements showed 95% limits of agreement with manual annotation. These results show that these CNN models are promising tools for median nerve ultrasound imaging analysis.

The effects of umbrella handle shape and grip type on muscle activation and postural variability under windy conditions.

This study investigated how different shapes of umbrella handles and grip types influence muscle activation and postural variability under windy conditions. Seventeen adult participants were enrolled in this study, and different handle shapes (cylindrical, ellipsoidal, and triangular prism-shaped), grip types (four- and five-finger grip), and wind strengths were tested. Activation of the forearm and upper arm muscles was recorded using surface electromyography. Postural variability and ratings of (1) perceived difficulty of use and (2) perceived grip-posture variability were measured. The results indicated that activation of the finger flexor muscle increased under windy conditions, whereas postural variability was not affected. Weak wind (3.2-7.5 m/s) conditions caused greater perceived postural variability and activation of the wrist extensor muscle. The ellipsoidal handle shape had lower endpoint postural variability when held with a five-finger grip and also had lower perceived postural variability and difficulty compared to that with the cylindrical shape. Our findings can be used to select appropriate umbrella handle designs based on grip type and wind conditions.

Impact of task variation and microbreaks on muscle fatigue at seated and standing postures.

BACKGROUND: Prolonged and sustained work posture among computer users is one of the main factors that contributes to musculoskeletal discomfort. Rest-break interventions such as task variation and microbreaks may help prevent muscle fatigue and work-related musculoskeletal disorder. OBJECTIVE: We aimed to investigate the effects of task variation and microbreaks at seated and standing workstations on forearm muscle activity, namely extensor digitorum communis, extensor carpi ulnaris, flexor digitorum superficialis, and flexor carpi ulnaris; mouse operation force (vertical compression force); mouse operation parameters; and perceived body discomfort during mouse operation. METHODS: Twelve healthy right-handed young adults were recruited (male: n = 7, 21.6±1.4 years; female: n = 5, 21.4±1.7 years). Participants performed three blocks of computer tasks (computer mouse operation and typing) in both seated and standing postures with each block lasting for 30 min. Surface electromyography (EMG) of the forearm muscles and operation force were monitored during computer mouse operation. Body discomfort rating was recorded at the end of each block. RESULTS: With simulated task variation and microbreaks, work posture and work time showed no significant difference with EMG amplitude and mouse operation force. CONCLUSION: Task variation and microbreaks could be of benefit to computer users by reducing muscle fatigue during long hours of computer work at both seated and standing workstations.

Users' adaptations to the proportional speed control of a motorised walker.

PURPOSE: Speed control is commonly used to regulate the forces applied by motorised walkers (MW) and there are often situations where the speed targeted deviates from the preferred walking speed of its users, such as when encouraging higher walking speeds and due to safety consideration. This study investigates the effects of different MW's target speeds on the selected walking speeds, force applied, perceived exertion, and gait of MW users during steady-state walking. MATERIALS AND METHODS: The spatiotemporal gait parameters and perceived exertion of twenty young healthy participants were measured as they walked at a comfortable, self-selected speed using a MW as it was controlled to target forward speeds of 0.6, 0.8, 1.0, 1.2, and 1.4 m s-1 as well as when no assistive force was applied by the MW. RESULTS: On average, users would walk slower when their "No Assist" walking speed is higher than the MW's speed target and vice versa. Additionally, the force applied to the MW is proportional to the difference in speed, either faster or slower, when compared to "No Assist". CONCLUSION: The user's exertion and the energy used by the MW are both minimised when target speed is close to the preferred walking speed of the user. Additionally, these findings suggest that the speed target can be used to change the walking speed of users but only to a certain extend and at the cost of higher perceived exertion.Implications for rehabilitationThe larger the difference between the target speed of the MW and the preferred walking speed of the user, the more likely the user is to push or pull on the MW.Users would push or pull on the MW with a force proportional to the difference from their preferred walking speed even when matching the MW's target speed.Users can be encouraged to walk at higher than preferred speeds, even though this would come at the cost of higher perceived exertion.

Effects of Simulated Visual Impairment Conditions on Movement and Anxiety during Gap Crossing.

This study investigated the effects of visual conditions associated with progressive eye disease on movement patterns and anxiety levels during gap-crossing tasks. Notably, 15 healthy young adults performed crossover platforms with a 10 cm gap at three different heights, namely equal (0 cm), raised (+15 cm), and lowered (-15 cm) levels, under four vision conditions, namely normal or corrected eyesight, 10° tunnel vision, 5° tunnel vision, and 5° tunnel vision with 0.04 occlusion. Leg movements during gap crossing were analyzed using three-dimensional motion analysis. The results highlighted a distinct motion pattern in the trajectories of participants' legs under the different visual conditions. Specifically, at the point where the gap-crossing movement began (D1), the normal or corrected eyesight conditions resulted in further separation between the steps compared with the other visual conditions. The highest point of the foot during movement (D2) did not differ between the visual conditions, except for the 0 cm step. Furthermore, anxiety levels, as quantified by the State-Trait Anxiety Inventory (STAI-S) questionnaire, were exacerbated under conditions of restricted visual information. In conclusion, visual impairments associated with progressive ocular diseases may perturb complex motor movement patterns, including those involved in gap-crossing tasks, with heightened anxiety potentially amplifying these disturbances.

Backward Walking Styles and Impact on Spatiotemporal Gait Characteristics.

Forward walking (FW) is a common balance assessment tool. However, its sensitivity is limited by the ceiling effect. Reverse gait, such as backward walking (BW), has been reported to have more advantages than FW for balance assessment. Three factors related to postural instability (i.e., increased speeds, restricted arm swing, and reduced visual feedback) during BW were investigated to determine BW conditions that have the potential to predict falls. Three-dimensional analyses were used to analyze seven walking conditions. FW and BW at self-selected and fast speeds were analyzed to identify the effects of speed. Walking with normal arm swings, crossed arms, and abducted arms during BW was tested to determine the effects of arm position. BW with closed and open eyes was compared to investigate the effects of visual feedback. BW had a significantly shorter step length than FW at high speeds. When the arms were abducted, the stance phase (%) was significantly lower compared to when arms were crossed during BW. Moreover, BW with closed eyes revealed significantly higher mediolateral center of mass (COM) displacements than with open eyes. We observed that BW with fast speeds, a crossed arm position, and closed eyes has the potential to help assess fall risk because it requires higher balance ability through spatiotemporal and COM adjustment.

Differential associations between simple physical performance tests with global and specific cognitive functions in cognitively normal and mild cognitive impairment: a cross-sectional cohort study of Asian community-dwelling older adults.

BACKGROUND: Physical performance declines and executive dysfunctions are predictors of dementia. However, their associations are not well understood in Asian older adults without dementia (cognitively normal [CN] and mild cognitive impairment [MCI]), especially in a single study. OBJECTIVE: Examine the associations between physical performance measures with executive function (EF)-based and non-EF-based neurocognitive tests and whether preclinical dementia cognitive status i.e., CN and MCI, moderated these associations. METHODS: We examined cross-sectional cohort of 716 community-dwelling older adults without dementia (CN = 562 and MCI = 154) using multivariable linear regression models. We associated three simple physical performance measures, namely timed-up-and-go (TUG), fast gait speed (FGS), and 30-s chair stand test (30 s-CST), with a comprehensive neurocognitive test battery measuring EF and non-EF cognitive functions. Moderating effects of cognitive status on the associations were examined. In all models, we controlled for pertinent covariates, including age, education, medical and psychiatric status. RESULTS: Upon controlling for covariates, TUG was most strongly and positively associated with multiple EF-based neurocognitive tests, followed by FGS, with 30 s-CST having the weakest associations. For all physical performance measures, no significant associations with non-EF-based neurocognitive tests were detected. Cognitive status significantly moderated the associations between all physical measures and several neurocognitive tests, with stronger associations in the MCI than CN. CONCLUSION: Compared to FGS and 30 s-CST, TUG had the most robust associations with multiple EF-based cognitive functions. Given their differential associations with global and detailed neurocognitive tests and significant moderating effects of cognitive status, findings highlight a need to carefully consider the choices of simple physical performance tests when using these tests with a heterogenous group of community-dwelling older adults without dementia.

The Effects of Standing Working Posture on Operation Force and Upper Limb Muscle Activation When Using Different Pointing Devices.

This study investigated how sitting and standing working postures affected operation force, upper limb muscle activation, and task performance using different pointing devices. Fifteen male participants completed cursor aiming and dragging tasks using a conventional mouse, a vertical mouse, and a trackball at sitting and standing workstations. A custom-made force plate was used to measure operation forces applied to the pointing devices. Surface electromyography (EMG) was used to capture the activation of the biceps brachii, triceps brachii, deltoid, and trapezius. Task performance was measured by task success rates, and subjective ratings were obtained for the force required for operation, smoothness of operation, accuracy, and local fatigue in the upper limb. We quantified the following significant outcomes: (1) greater operation forces were found when standing; (2) standing reduced EMG amplitudes of the triceps and trapezius muscles for all tasks; (3) during the aiming task, the vertical mouse had greater operation forces; (4) during the dragging task, both the vertical mouse and trackball had greater operation forces; and (5) task success rates differed for pointing devices only when sitting. This study revealed the distinct biomechanical properties of standing working posture and suggested limited beneficial effects of alternative pointing devices in terms of task performance and subjective ratings.

Effects of various handle shapes and surface profiles on the hand-arm responses and comfort during short-term exposure to handle vibration.

The magnitude of hand- (HTV) and wrist- (WTV) transmitted vibration can negatively impact upper limb responses even during short-term exposure. This study aimed to establish the effects of various handle-grip designs on the harmful impacts of vibration, sustained grip exertion, and unnatural posture. The primary focus was to investigate how using a handle grip and how three shapes with two surface profiles affect HTV, WTV, and forearm muscle activities during exposure. The secondary goal was to evaluate the immediate effects on fundamental hand functions, perceived discomfort/comfort, and perceived vibration level after exposure. The final objective was to assess which of the handle designs had the least harmful effects. Fourteen young male adults were recruited and asked to consistently grip a vibrating handle structure for 2 min while the primary parameters were recorded. Pre- and post-task measurements of secondary parameters were recorded on the six design conditions and one control condition (no handle grip). The study found that implementing a regular circular-smooth handle resulted in lower transmitted vibrations, leading to lower upper-limb discomfort, higher grip comfort, and lower perceived vibration. Additionally, shape significantly affected HTV, resulting in grip strength reduction, while surface profile did not influence transmitted vibrations but significantly impacted ring and small finger sensitivity, finger and hand discomfort, and grip comfort. Finally, forearm muscle activities were unaffected, and no significant interaction effects were observed. Circular handles also had the least negative impacts, and elliptic handles had the most negative impacts on the upper extremity because of the level of hand-handle contact stress and hand-grip effort. Meanwhile, the uneven distribution of vibration on the fingers and palm imposed by the rounded spikes on the patterned surface led to decreased finger sensitivity, higher discomfort, and lower grip comfort. Therefore, when machine operation involves moderate grip exertion, pronated forearm posture, and short-term handle vibration exposure, implementing a hard-solid handle with less hand-handle contact area, less grip effort, and even texture is recommended.

The effect of horizontal forces from a Smart Walker on gait and perceived exertion.

Increasingly, electric motors are being incorporated into wheeled walkers to implement various smart features to better assist their users physically. These modified walkers, known as Smart Walkers, use their electric motors to generate horizontal forces that can be used to reduce the physical load for walking, prevent falls and provide navigation support. However, these forces can also alter gait and may inadvertently increase the exertion of the users. This study aims to describe the effects of assistive and resistive horizontal forces (from -18.47 N to 27.70 N) from a Smart Walker on gait and perceived exertion of its users during steady-state walking. Self-selected comfortable walking speed, cadence, stride length, double support phase and ratings of perceived exertion (RPE) were significantly affected and different effects were found for resistive force, relatively low assistive force and high assistive force. With increasing force from -18.47 N to 0 N, RPE decreased and the users walked with lower double support time. From 0 N to 9.23 N, RPE continued to decrease to its lowest point while gait parameters remained constant. Further increasing force up to 27.70 N increased RPE and led to the users to choose to walk at higher speeds. This study demonstrates that users adapt their gait significantly to the forces applied and relatively high constant forces, whether assistive or resistive, will increase perceived exertion. Hence, these need to be carefully considered when developing Smart Walkers in order to provide safe and effective support to its users.

Motor control characteristics of upper limbs in response to assistive forces during bilateral tasks.

Most research on power assist suits (PASs) that concerned PAS-human interactions has used human physical reactions as criteria to evaluate the mechanical function, however, with minimal emphasis on human reactions in response to PASs. In this study, we focused on the physiological responses of the upper limbs including muscle activity of the biceps brachii and the triceps brachii, co-activation, force steadiness (CV) and rated perceived exertion (RPE) to various patterns of bilateral assistive force, such as unilateral assistance (L0% & R67% [% = percentage of workload force, L = left arm, R = right arm], L67% & R0%, L0% & R33%, L33% & R0%), symmetrical (L0% & R0%, L33% & R33%, L67% & R67%) and asymmetrical bilateral assistance (L33% & R67%, L67% & R33%), during bilateral isometric force-matching tasks. The results showed a similar muscular response of the two arms to bilateral assistive conditions, and the muscle activity of the arm that was being observed decreased only when the assistive force that applied on itself increased, indicating that both arms may have adopted similar but independent motor control mechanisms to acclimate to the bilateral assistive forces. Comparison between the two unilateral assistances (L0% & R33% and L33% & R0%) and the two asymmetrical bilateral assistances (L33% & R67%, L67% & R33%) showed no significant differences in muscular responses, CV and RPE, indicating that bilateral assistances with bilateral interchanged assistive levels may be equally effective regardless of which arm the higher assistive force is applied to. Comparison between unilateral and symmetrical assistive conditions that have similar overall workloads (L67% & R0%, L33% & R33%, L0% & R67%) showed a lower CV and RPE score at symmetrical assistance compared with unilateral assistance, suggesting that assisting both arms with the same level simultaneously improves task performances compared with applying the assistive force to only one arm.

Changes in Muscle Activity in Response to Assistive Force during Isometric Elbow Flexion.

This study investigated the muscle activity and force variability in response to perturbation of assistive force during isometric elbow flexion. Sixteen healthy right-handed young men (age: 22.0 ± 1.1 years; height: 171.9 ± 4.8 cm; weight 68.4 ± 11.2 kg) were recruited and the muscle activity of biceps brachii and triceps brachii were assessed using surface electromyography. Workload force and assistive force applied on isometric elbow flexion significantly affected the changes in both biceps and triceps muscle activities. A higher assistive force was shown to result in reduced biceps muscle activity compared to the unassisted period. In contrast, the efficiency of the assistive force acting on the biceps decreased as the assistive force increased. In general, the force variability of the biceps muscle remained approximately the same at lower workload force conditions than that at higher workload force conditions. In conclusion, higher assistive force may not yield a higher performance efficiency in human-assistive force interaction.

Effects of mechanical assistance on muscle activity and motor performance during isometric elbow flexion.

Mechanical assistance on joint movement is generally beneficial; however, its effects on cooperative performance and muscle activity needs to be further explored. This study examined how motor performance and muscle activity are altered when mechanical assistance is provided during isometric force control of ramp-down and hold phases. Thirteen right-handed participants (age: 24.7 ± 1.8 years) performed trajectory tracking tasks. Participants were asked to maintain the reference magnitude of 47 N (REF) during isometric elbow flexion. The force was released to a step-down magnitude of either 75% REF or 50% REF and maintained, with and without mechanical assistance. The ramp-down durations of force release were set to 0.5, 2.5, or 5.0 s. Throughout the experiment, we measured the following: (1) the force output using load cells to compute force variability and overshoot ratio; (2) peak perturbation on the elbow movement using an accelerometer; (3) the surface electromyography (sEMG) from biceps brachii and triceps brachii muscles; and (4) EMG oscillation from the biceps brachii muscle in the bandwidth of 15-45 Hz. Our results indicated that mechanical assistance, which involved greater peak perturbation, demonstrated lower force variability than non-assistance (p < 0.01), while EMG oscillation in the biceps brachii muscle from 15 to 45 Hz was increased (p < 0.05). These findings imply that if assistive force is provided during isometric force control, the central nervous system actively tries to stabilize motor performance by controlling specific motor unit activity in the agonist muscle.

Force and electromyography responses during isometric force release of different rates and step-down magnitudes.

This study investigated motor responses of force release during isometric elbow flexion by comparing effects of different ramp durations and step-down magnitudes. Twelve right-handed participants (age: 23.1 ±â€¯1.1) performed trajectory tracking tasks. Participants were instructed to release their force from the reference magnitude (REF; 40% of maximal voluntary contraction force) to a step-down magnitude of 67% REF or 33% REF and maintain the released magnitude. Force release was guided by ramp durations of either 1 s or 5 s. Electromyography of the biceps brachii and triceps brachii was performed during the experimental task, and the co-contraction ratio was evaluated. Force output was recorded to evaluate the parameters of motor performance, such as force variability and overshoot ratio. Although a longer ramp duration of 5 s decreased the force variability and overshoot ratio than did shorter ramp duration of 1 s, higher perceived exertion and co-contraction ratio were followed. Force variability was greater when force was released to the step-down magnitude of 33% REF than that when the magnitude was 67% REF, however, the overshoot ratio showed opposite results. This study provided evidence proving that motor control strategies adopted for force release were affected by both duration and step-down magnitude. In particular, it implies that different control strategies are required according to the level of step-down magnitude with a relatively short ramp duration.

An overview of hand postures and aging on morphological changes of the median nerve.

BACKGROUND: High-resolution ultrasound is being widely used in carpal tunnel examination to understand morphological and biomechanical characteristics of the median nerve and surrounding anatomy structures. MAIN BODY: Healthy young and elderly men were recruited. The median nerve at proximal wrist region was examined by ultrasound imaging technique. A total of seven wrist angle was examined. Generally, the median nerve cross-sectional area of the elderly group is significantly larger than the young group. SHORT CONCLUSION: Wrist posture in greater flexion or extension caused a larger decrease in the median nerve cross-sectional area across both groups.

Deformation of the median nerve at different finger postures and wrist angles.

BACKGROUND: The objective of this study was to evaluate the changes of the median nerve cross-sectional area (MNCSA) and diameters of the median nerve at different finger postures and wrist angles. METHODS: Twenty-five healthy male participants were recruited in this study. The median nerve at wrist crease was examined at six finger postures, and repeated with the wrist in 30° flexion, neutral (0°), and 30° extension. The six finger postures are relaxed, straight finger, hook, full fist, tabletop, and straight fist. RESULTS: The main effects of both finger postures and wrist angles are significant (p < 0.05) on changes of the MNCSA. Different finger tendon gliding postures cause a change in the MNCSA. Furthermore, wrist flexion and extension cause higher deformation of the MNCSA at different finger postures. DISCUSSION: The median nerve parameters such as MNCSA and diameter were altered by a change in wrist angle and finger posture. The results may help to understand the direct biomechanical stresses on the median nerve by different wrist-finger activities.

Effects of grip force on median nerve deformation at different wrist angles.

The present study investigated the effects of grip on changes in the median nerve cross-sectional area (MNCSA) and median nerve diameter in the radial-ulnar direction (D1) and dorsal-palmar direction (D2) at three wrist angles. Twenty-nine healthy participants (19 men [mean age, 24.2 ± 1.6 years]; 10 women [mean age, 24.0 ± 1.6 years]) were recruited. The median nerve was examined at the proximal carpal tunnel region in three grip conditions, namely finger relaxation, unclenched fist, and clenched fist. Ultrasound examinations were performed in the neutral wrist position (0°), at 30°wrist flexion, and at 30°wrist extension for both wrists. The grip condition and wrist angle showed significant main effects (p < 0.01) on the changes in the MNCSA, D1, and D2. Furthermore, significant interactions (p < 0.01) were found between the grip condition and wrist angle for the MNCSA, D1, and D2. In the neutral wrist position (0°), significant reductions in the MNCSA, D1, and D2 were observed when finger relaxation changed to unclenched fist and clenched fist conditions. Clenched fist condition caused the highest deformations in the median nerve measurements (MNCSA, approximately -25%; D1, -13%; D2, -12%). The MNCSA was significantly lower at 30°wrist flexion and 30°wrist extension than in the neutral wrist position (0°) at unclenched fist and clenched fist conditions. Notably, clenched fist condition at 30°wrist flexion showed the highest reduction of the MNCSA (-29%). In addition, 30°wrist flexion resulted in a lower D1 at clenched fist condition. In contrast, 30°wrist extension resulted in a lower D2 at both unclenched fist and clenched fist conditions. Our results suggest that unclenched fist and clenched fist conditions cause reductions in the MNCSA, D1, and D2. More importantly, unclenched fist and clenched fist conditions at 30°wrist flexion and 30°wrist extension can lead to further deformation of the median nerve.

An exploratory investigation of changes in gait parameters with age in elderly Japanese women.

BACKGROUND: The aim of the present study was to identify gait parameters in elderly Japanese women. 30 elderly women (65-74.9 years old) and 19 very elderly women (≥75 years old) participated in this study. A 3-dimensional (3D) motion analysis system was used to collect kinematic data, and a total of 70 gait parameters were analysed. Gait parameters included basic gait parameters, gait cycle parameters, and joints angle parameters, as well as angular velocity parameters, such as peak velocity and timing at the hip, knee, and ankle joints. RESULTS: Our results indicated that basic gait parameters, such as the gait cycle, peak joint angle timing, and angular velocity parameters, significantly differ between elderly and very elderly women. Delayed peak joint angle timing and angular velocity parameters occurred during critical phases throughout the gait cycle: pre-swing, initial swing, and terminal swing phases. CONCLUSIONS: Several gait parameters exhibited significant differences between elderly and very elderly women. The timing of the peak joint angle and angular velocity parameters are primary characteristics defining gait changes in the elderly.

Median nerve behavior at different wrist positions among older males.

The effect of wrist flexion-extension on the median nerve appearance, namely the cross-sectional area (MNCSA) and the longitudinal (D1) and vertical (D2) diameters, was investigated among older adults (N = 34). Ultrasound examination was conducted to examine the median nerve at different wrist angles (neutral; and 15°, 30°, and 45° extension and flexion), in both the dominant and nondominant hand. Median nerve behavior were significantly associated with wrist angle changes. The MNCSA at wrist flexion and extension were significantly smaller (P < .001) compared with the neutral position in both the dominant and nondominant hand. The D1 and D2 were significantly reduced at flexion (P < .001) and extension (P < .001), respectively, in both the dominant and nondominant hand. Our results suggest that a larger flexion-extension angle causes higher compression stress on the median nerve, leading to increased deformation of the MNCSA, D1, and D2 among older adults.

Effect of wrist angle on median nerve appearance at the proximal carpal tunnel.

This study investigated the effects of wrist angle, sex, and handedness on the changes in the median nerve cross-sectional area (MNCSA) and median nerve diameters, namely longitudinal diameter (D1) and vertical diameter (D2). Ultrasound examination was conducted to examine the median nerve at the proximal carpal tunnel in both dominant and nondominant hands of men (n = 27) and women (n = 26). A total of seven wrist angles were examined: neutral; 15°, 30°, and 45° extension; and 15°, 30°, and 45° flexion. Our results indicated sexual dimorphism and bilateral asymmetry of MNCSA, D1 and D2 measurements. MNCSA was significantly reduced when the wrist angle changed from neutral to flexion or extension positions. At flexion positions, D1 was significantly smaller than that at neutral. In contrast, at extension positions, D2 was significantly smaller than that at neutral. In conclusion, this study showed that MNCSA decreased as the wrist angle changed from neutral to flexion or extension positions in both dominant and nondominant hands of both sexes, whereas deformation of the median nerve differed between wrist flexion and extension.