The highs and lows of lifting loads: SPM analysis of multi-segmental spine angles in healthy adults during manual handling with increased load

Introduction: Manual handling personnel and those performing manual handling tasks in non-traditional manual handling industries continue to suffer debilitating and costly workplace injuries. Smart assistive devices are one solution to reducing musculoskeletal back injuries. Devices that provide targeted assistance need to be able to predict when and where to provide augmentation via predictive algorithms trained on functional datasets. The aim of this study was to describe how an increase in load impacts spine kinematics during a ground-to-platform manual handling task. Methods: Twenty-nine participants performed ground-to-platform lifts for six standardised loading conditions (50%, 60%, 70%, 80%, 90%, and 100% of maximum lift capacity). Six thoracic and lumbar spine segments were measured using inertial measurement units that were processed using an attitude-heading-reference filter and normalised to the duration of the lift. The lift was divided into four phases weight-acceptance, standing, lift-to-height and place-on-platform. Statistical significance of sagittal angles from the six spine segments were identified through statistical parametric mapping one-way analysis of variance with repeated measures and post hoc paired t-tests. Results: Two regions of interest were identified during a period of peak flexion and a period of peak extension. There was a significant increase in spine range of motion and peak extension angle for all spine segments when the load conditions were increased (p < 0.001). There was a decrease in spine angles (more flexion) during the weight acceptance to standing phase at the upper thoracic to upper lumbar spine segments for some condition comparisons. A significant increase in spine angles (more extension) during the place-on-platform phase was seen in all spine segments when comparing heavy loads (>80% maximum lift capacity, inclusive) to light loads (<80% maximum lift capacity) (p < 0.001). Discussion: The 50%-70% maximum lift capacity conditions being significantly different from heavier load conditions is representative that the kinematics of a lift do change consistently when a participant's load is increased. The understanding of how changes in loading are reflected in spine angles could inform the design of targeted assistance devices that can predict where and when in a task assistance may be needed, possibly reducing instances of back injuries in manual handling personnel.

A Green Synthesis Route to Derive Carbon Quantum Dots for Bioimaging Cancer Cells

Carbon quantum dots (CQDs) are known for their biocompatibility and versatile applications in the biomedical sector. These CQDs retain high solubility, robust chemical inertness, facile modification, and good resistance to photobleaching, which makes them ideal for cell bioimaging. Many fabrication processes produce CQDs, but most require expensive equipment, toxic chemicals, and a long processing time. This study developed a facile and rapid toasting method to prepare CQDs using various slices of bread as precursors without any additional chemicals. This fast and cost-effective toasting method could produce CQDs within 2 h, compared with the 10 h process in the commonly used hydrothermal method. The CQDs derived from the toasting method could be used to bioimage two types of colon cancer cells, namely, CT-26 and HT-29, derived from mice and humans, respectively. Significantly, these CQDs from the rapid toasting method produced equally bright images as CQDs derived from the hydrothermal method.

Exoskeleton Application to Military Manual Handling Tasks

OBJECTIVE

The aim of this review was to determine how exoskeletons could assist Australian Defence Force personnel with manual handling tasks.

BACKGROUND

Musculoskeletal injuries due to manual handling are physically damaging to personnel and financially costly to the Australian Defence Force. Exoskeletons may minimize injury risk by supporting, augmenting, and/or amplifying the user's physical abilities. Exoskeletons are therefore of interest in determining how they could support the unique needs of military manual handling personnel.

METHOD

Industrial and military exoskeleton studies from 1990 to 2019 were identified in the literature. This included 67 unique exoskeletons, for which Information about their current state of development was tabulated.

RESULTS

Exoskeleton support of manual handling tasks is largely through squat/deadlift (lower limb) systems (64%), with the proposed use case for these being load carrying (42%) and 78% of exoskeletons being active. Human-exoskeleton analysis was the most prevalent form of evaluation (68%) with reported reductions in back muscle activation of 15%-54%.

CONCLUSION

The high frequency of citations of exoskeletons targeting load carrying reflects the need for devices that can support manual handling workers. Exoskeleton evaluation procedures varied across studies making comparisons difficult. The unique considerations for military applications, such as heavy external loads and load asymmetry, suggest that a significant adaptation to current technology or customized military-specific devices would be required for the introduction of exoskeletons into a military setting.

APPLICATION

Exoskeletons in the literature and their potential to be adapted for application to military manual handling tasks are presented.

Carbon Dot Therapeutic Platforms: Administration, Distribution, Metabolism, Excretion, Toxicity, and Therapeutic Potential

Ultrasmall nanoparticles are often grouped under the broad umbrella term of "nanoparticles" when reported in the literature. However, for biomedical applications, their small sizes give them intimate interactions with biological species and endow them with unique functional physiochemical properties. Carbon quantum dots (CQDs) are an emerging class of ultrasmall nanoparticles which have demonstrated considerable biocompatibility and have been employed as potent theragnostic platforms. These particles find application for increasing drug solubility and targeting, along with facilitating the passage of drugs across impermeable membranes (i.e., blood brain barrier). Further functionality can be triggered by various environmental conditions or external stimuli (i.e., pH, temperature, near Infrared (NIR) light, ultrasound), and their intrinsic fluorescence is valuable for diagnostic applications. The focus of this review is to shed light on the therapeutic potential of CQDs and identify how they travel through the body, reach their site of action, administer therapeutic effect, and are excreted. Investigation into their toxicity and compatibility with larger nanoparticle carriers is also examined. The future of CQDs for theragnostic applications is promising due to their multifunctional attributes and documented biocompatibility. As nanomaterial platforms become more commonplace in clinical treatments, the commercialization of CQD therapeutics is anticipated.

Prediction of gait trajectories based on the Long Short Term Memory neural networks

The forecasting of lower limb trajectories can improve the operation of assistive devices and minimise the risk of tripping and balance loss. The aim of this work was to examine four Long Short Term Memory (LSTM) neural network architectures (Vanilla, Stacked, Bidirectional and Autoencoders) in predicting the future trajectories of lower limb kinematics, i.e. Angular Velocity (AV) and Linear Acceleration (LA). Kinematics data of foot, shank and thigh (LA and AV) were collected from 13 male and 3 female participants (28 ± 4 years old, 1.72 ± 0.07 m in height, 66 ± 10 kg in mass) who walked for 10 minutes at preferred walking speed (4.34 ± 0.43 km.h-1) and at an imposed speed (5km.h-1, 15.4% ± 7.6% faster) on a 0% gradient treadmill. The sliding window technique was adopted for training and testing the LSTM models with total kinematics time-series data of 10,500 strides. Results based on leave-one-out cross validation, suggested that the LSTM autoencoders is the top predictor of the lower limb kinematics trajectories (i.e. up to 0.1s). The normalised mean squared error was evaluated on trajectory predictions at each time-step and it obtained 2.82-5.31% for the LSTM autoencoders. The ability to predict future lower limb motions may have a wide range of applications including the design and control of bionics allowing improved human-machine interface and mitigating the risk of falls and balance loss.

Soft gold nanowire sponge antenna for battery-free wireless pressure sensors

The past decade has witnessed growing interest in developing soft wearable pressure sensors with the ultimate goal of transforming today's hospital-centered diagnosis to tomorrow's patient-centered bio-diagnosis. In this context, battery-free wireless antenna-based pressure sensors will be highly advantageous for ubiquitous real-time health monitoring. However, current wireless antennas are largely based on thin films from traditional bulk metallic films or novel nanomaterials with an air-cavity design, which can only be operated in a limited pressure range due to the rigidity of active films and/or inherent cavity dimensions. Herein we report a soft battery-free wireless pressure sensor that is based on a three-dimensional (3D) porous gold nanowire foam-elastomer composite and is fabricated by solution-based conformal electroless plating technology, followed by elastomer encapsulation. We observe a transducer trade-off point for our foam antenna, below which the inductive effect and capacitive effect function together and above which the capacitive effect dominates. When an external pressure is applied, initially the inductance and capacitance increase simultaneously but the capacitance decreases afterwards. This can be transformed into a variable resonant frequency that first decreases linearly and then increases (in the capacitance domination pressure range). Importantly, the linear detection range of the sensor can be tuned simply by adjusting the thickness of the sponge or the rigidity of the elastomer (PDMS). We can achieve a wide pressure range of 0-248 kPa, which is the largest linear detection range reported in the literature (typically from 0 to 30 kPa) to the best of our knowledge. As a proof of concept, we further demonstrated that our gold nanowire foam sensor can be used to weigh people under both static and dynamic conditions.

A real-time fuzzy logic biofeedback controller for freestyle swimming body posture adjustment

Wearable body area networks (BANs) have been widely used in activity measurements for kinematic information collection. This paper presents the design and implementation of a wearable device used as a training tool in freestyle swimming. The device supplies a close-loop control mechanism via a fuzzy logic controller. Swimming posture data is collected quantitatively and audibly fed back to swimmers in real time through bone conductors. Two recreational swimmers were invited to participate in a series of experiments including 7 days of baseline capability test (no feedback), 7 days of feedback training, and 2 days of retention test. It was found that both swimmers could well adapt to the feedback instructions. A maximum of 7.62% of lap time improvement and 29.64% of trunk roll improvement were observed in FB training, and such pattern was maintained after feedback was removed. We conclude that real-time fuzzy logic feedback can be used to improve recreational swimmers performance.

Bioimaging of C2C12 Muscle Myoblasts Using Fluorescent Carbon Quantum Dots Synthesized from Bread

Biocompatible carbon quantum dots (CQDs) have recently attracted increased interest in biomedical imaging owing to their advantageous photoluminescence properties. Numerous precursors of fluorescent CQDs and various fabrication procedures are also reported in the literature. However; the use of concentrated mineral acids and other corrosive chemicals during the fabrication process curtails their biocompatibility and severely limits the utilization of the products in cell bio-imaging. In this study; a facile; fast; and cost-effective synthetic route is employed to fabricate CQDs from a natural organic resource; namely bread; where the use of any toxic chemicals is eliminated. Thus; the novel chemical-free technique facilitated the production of luminescent CQDs that were endowed with low cytotoxicity and; therefore; suitable candidates for bioimaging sensors. The above mentioned amorphous CQDs also exhibited fluorescence over 360-420 nm excitation wavelengths; and with a broad emission range of 360-600 nm. We have also shown that the CQDs were well internalized by muscle myoblasts (C2C12) and differentiated myotubes; the cell lines which have not been reported before.

Lower Limb Kinematics Trajectory Prediction Using Long Short-Term Memory Neural Networks

This study determined whether the kinematics of lower limb trajectories during walking could be extrapolated using long short-term memory (LSTM) neural networks. It was hypothesised that LSTM auto encoders could reliably forecast multiple time-step trajectories of the lower limb kinematics, specifically linear acceleration (LA) and angular velocity (AV). Using 3D motion capture, lower limb position-time coordinates were sampled (100 Hz) from six male participants (age 22 ± 2 years, height 1.77 ± 0.02 m, body mass 82 ± 4 kg) who walked for 10 min at 5 km/h on a 0% gradient motor-driven treadmill. These data were fed into an LSTM model with a sliding window of four kinematic variables with 25 samples or time steps: LA and AV for thigh and shank. The LSTM was tested to forecast five samples (i.e., time steps) of the four kinematic input variables. To attain generalisation, the model was trained on a dataset of 2,665 strides from five participants and evaluated on a test set of 1 stride from a sixth participant. The LSTM model learned the lower limb kinematic trajectories using the training samples and tested for generalisation across participants. The forecasting horizon suggested higher model reliability in predicting earlier future trajectories. The mean absolute error (MAE) was evaluated on each variable across the single tested stride, and for the five-sample forecast, it obtained 0.047 m/s2 thigh LA, 0.047 m/s2 shank LA, 0.028 deg/s thigh AV and 0.024 deg/s shank AV. All predicted trajectories were highly correlated with the measured trajectories, with correlation coefficients greater than 0.98. The motion prediction model may have a wide range of applications, such as mitigating the risk of falls or balance loss and improving the human-machine interface for wearable assistive devices.

Local Crack-Programmed Gold Nanowire Electronic Skin Tattoos for In-Plane Multisensor Integration

Sensitive, specific, yet multifunctional tattoo-like electronics are ideal wearable systems for "any time, any where" health monitoring because they can virtually become parts of the human skin, offering a burdenless "unfeelable" wearing experience. A skin-like, multifunctional electronic tattoo made entirely from gold using a standing enokitake-mushroom-like vertically aligned nanowire membrane in conjunction with a programmable local cracking technology is reported. Unlike previous multifunctional systems, only a single material type is needed for the integrated gold circuits involved in interconnects and multiplexed specific sensors, thereby avoiding the use of complex multimaterials interfaces. This is possiblebecause the programmable local cracking technology allows for the arbitrary fine-tuning of the properties of elastic gold conductors from strain-insensitive to highly strain-sensitive simply by adjusting localized crack size, shape, and orientations-a capability impossible to achieve with previous bulk cracking technology. Furthermore, in-plane integration of strain/pressure sensors, anisotropic orientation-specific sensors, strain-insensitive stretchable interconnects, temperature sensors, glucose sensors, and lactate sensors without the need of soldering or gluing are demonstrated. This strategy opens a new general route for the design of next-generation wearable electronic tattoos.

A Review of Implant Communication Technology in WBAN: Progress and Challenges

Over the past six decades, there has been tremendous progress made in the field of medical implant communications. A comprehensive review of the progress, current state of the art, and future direction is presented in this paper. Implanted medical devices (IMDs) are designed mainly for the purpose of diagnostic, therapeutic, and assistive applications in heathcare, active living, and sports technology. The primary target of IMDs' design revolves around reliable communications, sustainable power sources, and a high degree of miniaturization while maintaining biocompatibility to surrounding tissues adhering to the human safety limits set by appropriate guidelines. The role of the Internet of Things and intelligent data analysis in implant device networks as future research is presented in this paper. Finally, in addition to reviewing the state of the art, a novel intuitive lower bound on implant size is presented.

A Survey on Biofeedback and Actuation in Wireless Body Area Networks (WBANs)

Wireless body area networks (WBANs) have attained increasing popularity as the next generation framework of wearable technologies for human monitoring. Invasive or noninvasive wearable sensors designed in a WBAN are worn to gather vital information. Biofeedback is a recent concept where collected data are used to generate actuation signals in WBANs. Applications can be seen in various areas such as sports (e.g., locomotor velocity) or medicine (e.g., blood pressure measurement). However, since the body is closely regulated, the next generation WBAN technology must be smart enough to react to monitored data. The main aim of this paper is to review the current state of biofeedback and actuation technology on WBANs in terms of its structure, applications, benefits, and control approaches. The emphasis on the specific requirements when applying biofeedback to humans will be highlighted and discussed. Challenges and open research issues will be concluded at the end.

Characterization of Impulse Radio Intrabody Communication System for Wireless Body Area Networks

Intrabody communication (IBC) is a promising data communication technique for body area networks. This short-distance communication approach uses human body tissue as the medium of signal propagation. IBC is defined as one of the physical layers for the new IEEE 802.15.6 or wireless body area network (WBAN) standard, which can provide a suitable data rate for real-time physiological data communication while consuming lower power compared to that of radio-frequency protocols such as Bluetooth. In this paper, impulse radio (IR) IBC (IR-IBC) is examined using a field-programmable gate array (FPGA) implementation of an IBC system. A carrier-free pulse position modulation (PPM) scheme is implemented using an IBC transmitter in an FPGA board. PPM is a modulation technique that uses time-based pulse characteristics to encode data based on IR concepts. The transmission performance of the scheme was evaluated through signal propagation measurements of the human arm using 4- and 8-PPM transmitters, respectively. 4 or 8 is the number of symbols during modulations. It was found that the received signal-to-noise ratio (SNR) decreases approximately 8.0 dB for a range of arm distances (5–50 cm) between the transmitter and receiver electrodes with constant noise power and various signal amplitudes. The SNR for the 4-PPM scheme is approximately 2 dB higher than that for the 8-PPM one. In addition, the bit error rate (BER) is theoretically analyzed for the human body channel with additive white Gaussian noise. The 4- and 8-PPM IBC systems have average BER values of 10⁻⁵ and 10⁻¹⁰, respectively. The results indicate the superiority of the 8-PPM scheme compared to the 4-PPM one when implementing the IBC system. The performance evaluation of the proposed IBC system will improve further IBC transceiver design.

Ultrasensitive Strain Sensor Produced by Direct Patterning of Liquid Crystals of Graphene Oxide on a Flexible Substrate

Ultrasensitive flexible strain sensors were developed through the combination of shear alignment of a high concentration graphene oxide (GO) dispersion with fast and precise patterning of multiple rectangular features on a flexible substrate. Resistive changes in the reduced GO films were investigated under various uniaxial strain cycles ranging from 0.025 to 2%, controlled with a motorized nanopositioning stage. The devices uniquely combine a very small detection limit (0.025%) and a high gauge factor with a rapid fabrication process conducive to batch production.

A wearable biofeedback control system based body area network for freestyle swimming

Wearable posture measurement units are capable of enabling real-time performance evaluation and providing feedback to end users. This paper presents a wearable feedback prototype designed for freestyle swimming with focus on trunk rotation measurement. The system consists of a nine-degree-of-freedom inertial sensor, which is built in a central data collection and processing unit, and two vibration motors for delivering real-time feedback. Theses devices form a fundamental body area network (BAN). In the experiment setup, four recreational swimmers were asked to do two sets of 4 x 25m freestyle swimming without and with feedback provided respectively. Results showed that real-time biofeedback mechanism improves swimmers kinematic performance by an average of 4.5% reduction in session time. Swimmers can gradually adapt to feedback signals, and the biofeedback control system can be employed in swimmers daily training for fitness maintenance.

A Circuit Model of Real Time Human Body Hydration

Changes in human body hydration leading to excess fluid losses or overload affects the body fluid's ability to provide the necessary support for healthy living. We propose a time-dependent circuit model of real-time human body hydration, which models the human body tissue as a signal transmission medium. The circuit model predicts the attenuation of a propagating electrical signal. Hydration rates are modeled by a time constant τ, which characterizes the individual specific metabolic function of the body part measured. We define a surrogate human body anthropometric parameter θ by the muscle-fat ratio and comparing it with the body mass index (BMI), we find theoretically, the rate of hydration varying from 1.73 dB/min, for high θ and low τ to 0.05 dB/min for low θ and high τ. We compare these theoretical values with empirical measurements and show that real-time changes in human body hydration can be observed by measuring signal attenuation. We took empirical measurements using a vector network analyzer and obtained different hydration rates for various BMI, ranging from 0.6 dB/min for 22.7 [Formula: see text] down to 0.04 dB/min for 41.2 [Formula: see text]. We conclude that the galvanic coupling circuit model can predict changes in the volume of the body fluid, which are essential in diagnosing and monitoring treatment of body fluid disorder. Individuals with high BMI would have higher time-dependent biological characteristic, lower metabolic rate, and lower rate of hydration.

Tattoolike Polyaniline Microparticle-Doped Gold Nanowire Patches as Highly Durable Wearable Sensors

Wearable and highly sensitive strain sensors are essential components of electronic skin for future biomonitoring and human machine interfaces. Here we report a low-cost yet efficient strategy to dope polyaniline microparticles into gold nanowire (AuNW) films, leading to 10 times enhancement in conductivity and ∼8 times improvement in sensitivity. Simultaneously, tattoolike wearable sensors could be fabricated simply by a direct "draw-on" strategy with a Chinese penbrush. The stretchability of the sensors could be enhanced from 99.7% to 149.6% by designing curved tattoo with different radius of curvatures. We also demonstrated roller coating method to encapusulate AuNWs sensors, exhibiting excellent water resistibility and durability. Because of improved conductivity of our sensors, they can directly interface with existing wireless circuitry, allowing for fabrication of wireless flexion sensors for a human finger-controlled robotic arm system.

Minimum toe clearance events in divided attention treadmill walking in older and young adults: a cross-sectional study

Background

Falls in older adults during walking frequently occur while performing a concurrent task; that is, dividing attention to respond to other demands in the environment. A particularly hazardous fall-related event is tripping due to toe-ground contact during the swing phase of the gait cycle. The aim of this experiment was to determine the effects of divided attention on tripping risk by investigating the gait cycle event Minimum Toe Clearance (MTC).

Methods

Fifteen older adults (mean 73.1 years) and 15 young controls (mean 26.1 years) performed three walking tasks on motorized treadmill: (i) at preferred walking speed (preferred walking), (ii) while carrying a glass of water at a comfortable walking speed (dual task walking), and (iii) speed-matched control walking without the glass of water (control walking). Position-time coordinates of the toe were acquired using a 3 dimensional motion capture system (Optotrak NDI, Canada). When MTC was present, toe height at MTC (MTC_Height) and MTC timing (MTC_Time) were calculated. The proportion of non-MTC gait cycles was computed and for non-MTC gait cycles, toe-height was extracted at the mean MTC_Time.

Results

Both groups maintained mean MTC_Height across all three conditions. Despite greater MTC_Height SD in preferred gait, the older group reduced their variability to match the young group in dual task walking. Compared to preferred speed walking, both groups attained MTC earlier in dual task and control conditions. The older group’s MTC_Time SD was greater across all conditions; in dual task walking, however, they approximated the young group’s SD. Non-MTC gait cycles were more frequent in the older group across walking conditions (for example, in preferred walking: young – 2.9 %; older - 18.7 %).

Conclusions

In response to increased attention demands older adults preserve MTC_Height but exercise greater control of the critical MTC event by reducing variability in both MTC_Height and MTC_Time. A further adaptive locomotor control strategy to reduce the likelihood of toe-ground contacts is to attain higher mid-swing clearance by eliminating the MTC event, i.e. demonstrating non-MTC gaits cycles.

Investigation of galvanic-coupled intrabody communication using the human body circuit model

Intrabody Communication (IBC) is a technique that uses the human body as a transmission medium for electrical signals to connect wearable electronic sensors and devices. Understanding the human body as the transmission medium in IBC paves way for practical implementation of IBC in body sensor networks. In this study, we propose a model for galvanic coupling-type IBC based on a simplified equivalent circuit representation of the human upper arm. We propose a new way to calculate the electrode-skin contact impedance. Based on the model and human experimental results, we discuss important characteristics of galvanic coupling-type IBC, namely, the effect of tissues, anthropometry of subjects, and electrode configuration on signal propagation. We found that the dielectric properties of the muscle primarily characterize the received signal when receiver electrodes are located close to transmitter electrodes. When receiver and transmitter electrodes are far apart, the skin dielectric property affects the received signal.

Limb joint effects on signal transmission in capacitive coupled intra-body communication systems

This paper contributes empirical measurements towards an understanding of signal attenuation in intra-body communication (IBC) systems due to limb posture effects. Recent studies have shown a degradation of transmission signals for IBC transmissions between limb segments, but these degradations have yet to be quantified with respect to relative limb position and within the transmission frequency range from 300 KHz to 200 MHz. We examine the impact of limb position specifically the effect of elbow joint flexion and extension into account using a portable vector network analyzer. The results presented indicate that the signal attenuation is larger in the case of extension, i.e., when the angle between forearm and upper arm increases. The minimum attenuation was 20.64 dB and 24.81 dB for the fix distance of 15 cm between transmitter and receiver electrodes and the joint angle of 45 and 180 degree respectively. It was found that attenuation decreased at an approximately linear rate over 300 KHz to 100 MHz and increased over the frequency range from 100 MHz to 200 MHz for the input signal frequency range from 300 KHz to 200 MHz. It was concluded that the minimum attenuation for the range of flexions and extensions occurred in the range 80-100 MHz. Future work will explore theoretical models to explain the observed results.

A note on octonionic support vector regression

This note presents an analysis of the octonionic form of the division algebraic support vector regressor (SVR) first introduced by Shilton A detailed derivation of the dual form is given, and three conditions under which it is analogous to the quaternionic case are exhibited. It is shown that, in the general case of an octonionic-valued feature map, the usual "kernel trick" breaks down. The cause of this (and its interpretation) is discussed in some detail, along with potential ways of extending kernel methods to take advantage of the distinct features present in the general case. Finally, the octonionic SVR is applied to an example gait analysis problem, and its performance is compared to that of the least squares SVR, the Clifford SVR, and the multidimensional SVR.

Swing limb mechanics and minimum toe clearance in people with knee osteoarthritis

OBJECTIVE

Knee osteoarthritis (OA) has been shown to be a risk factor for falls. Reductions in foot clearance during the swing phase of walking can cause a trip and potentially lead to a fall. This study examined the swing phase mechanics of people with and without knee OA during walking.

DESIGN

Minimum toe clearance (MTC) height, joint angles at the time of MTC and the influence of the angular changes of the hip, knee and ankle of the swing leg on foot clearance using sensitivity analysis were investigated in 50 knee OA participants and 28 age-matched asymptomatic controls.

RESULTS

Although both groups had a similar MTC height (controls: 12.8±6.7 mm, knee OA: 13.4±7.0 mm), the knee OA group used a different strategy to achieve the same foot clearance, as evidenced by greater knee flexion (52.5±5.3° vs 49.4±4.8°, p=0.007), greater hip abduction (-3.6±3.3° vs -1.8±3.3°, p=0.03) and less ankle adduction (2.8±1.9° vs 4.2±2.1°, p=0.01).

CONCLUSION

MTC height was comparable between the groups, however a different swing phase mechanism was used by the knee OA. Although adequate MTC is an important component of safe locomotion, it does not appear to be impaired in people with knee OA. Other factors, such as inadequate responses to postural perturbation, may be responsible for falls in this group.

Estimation of end point foot clearance points from inertial sensor data

Foot clearance parameters provide useful insight into tripping risks during walking. This paper proposes a technique for the estimate of key foot clearance parameters using inertial sensor (accelerometers and gyroscopes) data. Fifteen features were extracted from raw inertial sensor measurements, and a regression model was used to estimate two key foot clearance parameters: First maximum vertical clearance (m x 1) after toe-off and the Minimum Toe Clearance (MTC) of the swing foot. Comparisons are made against measurements obtained using an optoelectronic motion capture system (Optotrak), at 4 different walking speeds. General Regression Neural Networks (GRNN) were used to estimate the desired parameters from the sensor features. Eight subjects foot clearance data were examined and a Leave-one-subject-out (LOSO) method was used to select the best model. The best average Root Mean Square Errors (RMSE) across all subjects obtained using all sensor features at the maximum speed for m x 1 was 5.32 mm and for MTC was 4.04 mm. Further application of a hill-climbing feature selection technique resulted in 0.54-21.93% improvement in RMSE and required fewer input features. The results demonstrated that using raw inertial sensor data with regression models and feature selection could accurately estimate key foot clearance parameters.

Ultrasound monitoring of inter-knee distances during gait

Knee osteoarthritis is an extremely common, debilitating disease associated with pain and loss of function. There is considerable interest in monitoring lower limb alignment due to its close association with joint overload leading to disease progression. The effects of gait modifications that can lower joint loading are of particular interest. Here we describe an ultrasound-based system for monitoring an important aspect of dynamic lower limb alignment, the inter-knee distance during walking. Monitoring this gait parameter should facilitate studies in reducing knee loading, a primary risk factor of knee osteoarthritis progression. The portable device is composed of an ultrasound sensor connected to an Intel iMote2 equipped with Bluetooth wireless capability. Static tests and calibration results show that the sensor possesses an effective beam envelope of 120 degrees, with maximum distance errors of 10% at the envelope edges. Dynamic walking trials reveal close correlation of inter-knee distance trends between that measured by an optical system (Optotrak Certus NDI) and the sensor device. The maximum average root mean square error was found to be 1.46 cm. Future work will focus on improving the accuracy of the device.

On the feasibility of learning to predict minimum toe clearance under different walking speeds

A major concern in human movement research is preventing tripping and falling which is known to cause severe injuries and high fatalities in elderly (>65 years) populations. Current falls prevention technology consists of active interventions e.g., strength and balance exercises, preimpact fall detectors, and passive interventions e.g., shower rails, hip protectors. However it has been found that these interventions with the exception of balance exercises do not effectively reduce falls risk. Recent work has shown that the minimum toe clearance (MTC) can be successfully monitored to detect gait patterns indicative of tripping and falling risk. In this paper, we investigate the feasibility of predicting MTC values of consecutive gait cycles under different walking speeds. The objective is two-fold, first to determine if end point foot trajectories can be accurately predicted and second, if walking speed is a significant parameter which influences the prediction process. The Generalized Regression Neural Networks and the Support Vector Regressor models were trained to predict MTC time series successively over an increasing prediction horizon i.e., 1 to 10 steps. Increased walking speeds resulted in increased MTC variability but no significant increase in mean MTC height. Root mean squared prediction errors ranged between 2.2-2.6mm or 10% of the mean values of the respective test data. The SVM slightly outperformed the GRNN predictions (0.5%-2.1% better accuracy). Best prediction accuracies decreased by 0.5mm for a doubling of walking speed i.e., from 2.5 km/h to 5.5 km/h. The results are encouraging because they demonstrate that the technique could be applied to forecasting low MTC values and provide new approaches to falls prevention technologies.

A study on band-pass filtering for calculating foot displacements from accelerometer and gyroscope sensors

As a promising alternative to laboratory-constrained video capture systems in studies of human movement, inertial sensors (accelerometers and gyroscopes) are recently gaining popularity. Secondary quantities such as velocity, displacement and joint angles can be calculated through integration of acceleration and angular velocities. It is broadly accepted that this procedure is significantly influenced by accumulative errors due to integration, arising from sensor noise, non-linearities, asymmetries, sensitivity variations and bias drifts. In this paper, we assess the effectiveness of applying band-pass filtering to raw inertial sensor data under the assumption that sensor drift errors occur in the low frequency spectrum. The normalized correlation coefficient rho of the Fast Fourier Transform (FFT) spectra corresponding to vertical toe acceleration from inertial sensors and from a video capture system as a function of digital band-pass filter parameters is compared. The Root Mean Square Error (RMSE) of the vertical toe displacement for 30 second walking windows is calculated for 2 healthy subjects over a range of 4 walking speeds. The lowest RMSE and highest cross correlation achieved for the slowest walking speed of 2.5Km/h was 3.06cm and 0.871 respectively, and 2.96cm and 0.952 for the fastest speed of 5.5Km/h.

A division algebraic framework for multidimensional support vector regression

In this paper, division algebras are proposed as an elegant basis upon which to extend support vector regression (SVR) to multidimensional targets. Using this framework, a multitarget SVR called epsilon(Z)-SVR is proposed based on an epsilon-insensitive loss function that is independent of the coordinate system or basis used. This is developed to dual form in a manner that is analogous to the standard epsilon-SVR. The epsilon(H)-SVR is compared and contrasted with the least-square SVR (LS-SVR), the Clifford SVR (C-SVR), and the multidimensional SVR (M-SVR). Three practical applications are considered: namely, 1) approximation of a complex-valued function; 2) chaotic time-series prediction in 3-D; and 3) communication channel equalization. Results show that the epsilon(H)-SVR performs significantly better than the C-SVR, the LS-SVR, and the M-SVR in terms of mean-squared error, outlier sensitivity, and support vector sparsity.

Automatic recognition of gait patterns exhibiting patellofemoral pain syndrome using a support vector machine approach

Patellofemoral pain syndrome (PFPS) is a common disorder that afflicts people across all age groups, and results in various degrees of knee pain. The diagnosis of PFPS is difficult since the exact biomechanical factors and the extent to which they are affected by the disorder are still unknown. Recent research has reported significant statistical differences in ground reaction forces (GRFs) and foot kinematics, which could be indicative of PFPS, but the interrelationship between many of these measures and the pathology have been absent so far. In this paper, we applied the support vector machines (SVMs) to detect PFPS gait based on 14 GRF and 16 foot kinematic features recorded from 27 subjects (14 healthy and 13 with PFPS). The influence of combined gait parameters on classification performance was investigated through the use of a feature-selection algorithm. The optimal feature set was then compared against the most statistically significant individual features (p < 0.05) found by previous study. Test results indicated that GRF features alone resulted in a higher leave-one-out (LOO) classification accuracy (85.15%) compared to 74.07% using only kinematic features. A hill-climbing feature-selection algorithm was applied to determine the subset of combined kinematic and kinetic features, which provided optimal classifier performance. This subset, which consists of six features (two from GRF and four from foot kinematic features), provided an improved LOO accuracy of 88.89% . The optimal feature set detected by the SVM, which best identified gait characteristics of PFPS, was found to be closely related to inferential statistical analysis with the added distinction that the SVM could potentially be deployed as an automated system for detecting gait changes in patients with PFPS.

Detecting trunk motion changes due to pregnancy using pattern recognition techniques

There are anatomical changes during pregnancy due to the increased and altered mass distribution in the trunk that could lead to changes in gait. There is little research, however, regarding adaptations in trunk motion with pregnancy. In this paper, we investigated the application of two pattern recognition techniques: support vector machine (SVM) and linear discriminant analysis (LDA) to detect differences in trunk kinematics, when walking, between women in late pregnancy and nulliparous (control) women. Test results indicate that the SVM can identify the trunk motion of pregnant women from their counterparts with a better accuracy compared to the LDA (71.43% vs 28.57% respectively). Furthermore, with a feature selection technique applied, the accuracy improved to 95.24% % using only 2 features namely the pelvic sagittal plane displacement and thoracic lateral tilt displacement at heel contact. The results suggest that for better detection of trunk motion changes in pregnant women, non-linear analysis may be required. The SVM was able to effectively differentiate pregnancy related trunk motion changes during a walking task which may indicate altered musculoskeletal loads with potential for injury or pain.

The application of support vector machines for detecting recovery from knee replacement surgery using spatio-temporal gait parameters

Knee osteoarthritis (OA) is one of the leading causes of disability among the elderly which, depending on severity, may require surgical intervention. Knee replacement surgery provides pain relief and improves physical function including gait. However gait dysfunction such as altered spatio-temporal measures may persist after the surgery. In this paper, we investigated the application of support vector machines (SVM) to classify gait patterns indicative of knee OA before surgery based on 12 spatio-temporal gait parameters and investigated whether SVMs could be used to predict gait improvement 2 and 12 months following knee replacement surgery. Test results for the pre-operative data indicated that the SVM could successfully identify individuals with OA gait from the healthy using all of the spatio-temporal parameters with a maximum leave one out accuracy of 100% for the training set and 88.89% for the test set. Findings indicated that three patients still had altered gait patterns 2 months post-knee replacement surgery, but all individuals showed improvement in gait 12 months following surgery. Consequently, the SVM detected improvement in gait function due to surgical intervention at 2 and 12 months following knee replacement which coincided with clinical assessment of the knee. This suggests that spatio-temporal parameters contain important discriminative information which may be used for the identification of pathological gait using an SVM classifier.

A machine learning approach to k-step look-ahead prediction of gait variables from acceleration data

This paper investigates the use of machine learning to predict a sensitive gait parameter based on acceleration information from previous gait cycles. We investigate a k-step look-ahead prediction which attempts to predict gait variable values based on acceleration information in the current gait cycle. The variable is the minimum toe clearance which has been demonstrated to be a sensitive falls risk predictor. Toe clearance data was collected under normal walking conditions and 9 features consisting of peak acceleration and their normalized occurrences times were extracted. A standard least squares estimator, a generalized regression neural network (GRNN) and a support vector regressor (SVR) were trained using 60% of the data to estimate the minimum toe clearance and the remaining 40% was used to validate the model. It was found that when the training data contained data from all subjects (inter-subject) the best GRNN model provided a root mean square error (RMSE) of 2.8 mm, the best SVR had RMSE of 2.7 mm while the standard least squares linear regression method obtained 3.3 mm. When the training and test data consisted of different subject examples (inter-subject) data, the linear SVR demonstrated superior generalization capability (RMSE=3.3 mm) compared to other competing models. Validation accuracies up to 5-step look-ahead predictions revealed robust performances for both GRNN and SVR models with no clear degradation in prediction accuracy.

Support Vector Machines for detecting recovery from knee replacement surgery using quantitative gait measures

Knee osteoarthritis (OA) is one of the leading causes of disability among the elderly which, depending on severity, may require surgical intervention. Knee replacement surgery provides pain relief and improves physical function including gait. Gait dysfunction such as altered spatio-temporal measures and gait asymmetry both pre- and post-surgery, however, may still persist after the surgery. In this paper, we investigated the application of Support Vector Machines (SVM) to classify gait patterns pertaining to knee OA before surgery based on spatio-temporal gait parameters and to investigate whether SVM can assess gait improvement at 2 months following knee replacement surgery. Test results indicate that the SVM can identify the OA gait from the healthy ones with a max leave one out (LOO) accuracy of 94.2%. When feature selection technique was applied, the accuracy improved to 97.1% using only 2 symmetry index features. Further, the post surgery test results by the SVM indicated 4 patients still had altered gait. This suggests that subject gait symmetry should be monitored closely after surgery to assess treatment outcomes and recovery.

Detection of tripping gait patterns in the elderly using autoregressive features and support vector machines

Elderly tripping falls cost billions annually in medical funds and result in high mortality rates often perpetrated by pulmonary embolism (internal bleeding) and infected fractures that do not heal well. In this paper, we propose an intelligent gait detection system (AR-SVM) for screening elderly individuals at risk of suffering tripping falls. The motivation of this system is to provide early detection of elderly gait reminiscent of tripping characteristics so that preventive measures could be administered. Our system is composed of two stages, a predictor model estimated by an autoregressive (AR) process and a support vector machine (SVM) classifier. The system input is a digital signal constructed from consecutive measurements of minimum toe clearance (MTC) representative of steady-state walking. The AR-SVM system was tested on 23 individuals (13 healthy and 10 having suffered at least one tripping fall in the past year) who each completed a minimum of 10 min of walking on a treadmill at a self-selected pace. In the first stage, a fourth order AR model required at least 64 MTC values to correctly detect all fallers and non-fallers. Detection was further improved to less than 1 min of walking when the model coefficients were used as input features to the SVM classifier. The system achieved a detection accuracy of 95.65% with the leave one out method using only 16 MTC samples, but was reduced to 69.57% when eight MTC samples were used. These results demonstrate a fast and efficient system requiring a small number of strides and only MTC measurements for accurate detection of tripping gait characteristics.

Detecting idiopathic toe-walking gait pattern from normal gait pattern using heel accelerometry data and Support Vector Machines

Toe walking is commonly seen in children with neurological symptoms such as cerebral palsy. However idiopathic toe walking (ITW) in children is considered to be habitual. ITW children are categorized as toe walkers without any neurological problems, however they walk with their foot plantar-flexed. These children often suffer poor sport performance leading to low exercise levels and the associated consequences. If the condition is not treated, the ITW children eventually develop abnormal gait pattern as adults and could suffer from postural problems. However, ITW gait is difficult to observe since children can modify their gait when made aware of it. Gait analysis using heel accelerometry data in ITW children could provide an objective and quantitative description of their toe walking and may thus be beneficial for observing ITW. In this paper, we propose a technique based on Support Vector Machines (SVM) to recognize ITW gait patterns using heel accelerometry data. Test results indicated that the SVM is able to identify ITW gait patterns with a maximum accuracy of 87.5% when a feature selection algorithm was applied.

A prototype wireless inertial-sensing device for measuring toe clearance

Tripping and slipping are serious health concerns for the elderly because they result in life threatening injuries i.e., fractures and high medical costs. Our recent work in detection of tripping gait patterns has demonstrated that minimum toe clearance (MTC) is a sensitive falls risk predictor. MTC measurement has previously been done in gait laboratories and on treadmills which potentially imposes controlled walking conditions. In this paper, we describe a prototype design of a wireless device for monitoring vertical toe clearance. The sensors consists of a tri-axis accelerometer and dual-axis gyroscope connected to Crossbow sensor motes for wireless data transmission. Sensor data are transmitted to a laptop and displayed on a Matlab graphic user interface (GUI). We have performed zero base and treadmill experiments to investigate sensor performance to environmental variations and compared the calculated toe clearance against measurements made by an Optotrak motion system. It was found that device outputs were approximately independent of small ambient temperature variations, had a reliable range of 20m indoors and 50m outdoors and a maximum transmission rate of 20 packets/s. Toe clearance measurements were found to follow the Optotrak measurement trend but could be improved further by dealing with double integration errors and improving data transmission rates.