Accelerometer Output Research Articles

Defibrillator shocks and their effect on objective asian patient outcomes: results of the painfree SST clinical trial

Abstract Background The implantable cardioverter defibrillator (ICD) reduces mortality in patients at risk of sudden cardiac death. However, the effect of ICD shock on device-measured activity is unknown. The aim of this analysis was to analyze the acute and long-term effects of ICD shock on objective behavioral data. Methods The PainFree SST study was a prospective, multicenter, global study that enrolled 2770 patients including 225 from from Asia (181 from Japan, 25 from India, and 19 from Malaysia). Patient physical activity was determined from the device built-in accelerometer. The device translates the accelerometer output into a count of active minutes, which is stored as a daily total value in device memory for a rolling window of 425 days. Obviously, activity is measured only after device implantation. Early after implant, activity is reduced due to the procedure and thus not representative. Therefore, no baseline can be reported for activity. The follow-up activity data was analyzed with a linear mixed model, with a random effect for patient and an auto-regressive correlation matrix for subsequent measurements of individual patients. Results Figure 1 shows average daily activity between 90 days before and 90 days after a shock. The dashed vertical line indicates the day of shock. Both patients from Asia and from other countries show a drop in activity after a shock with recovery in the subsequent 30-60 days. The curve for Asian patients is more variable than the curve for other countries due to the lower number of patients (there were 33 Asian patients with shocks contributing data). Figure 2 shows average daily activity before and after ATP therapy. Any post shock measurement is excluded. Contrary to the temporary reduction in activity around a shock (Figure 1), ATP therapies do not seem to have an impact on activity. In patients from the other countries, activity is significantly reduced when patients are hospitalized (-66.3 minutes/day, p<0.0001), but not in Asian patients (-5.1 minutes/day, p=0.71). Both in Asian and other countries, activity is significantly reduced after an ICD shock (Asian: -47.5 minutes/day, p=0.0002; Other countries: -21.5 minutes/day, p<0.0001). The effect of shock for Asian is greater than the other countries (p=0.054). In the month after the shock there is a significant recovery, +34.8 minutes/day in Asian patients (p=0.0077) and +7.7 minutes/day in patients from other countries (p=0.031). The recovery is trended to be bigger in Asian patients (p=0.058). Conclusions ICD shocks have a lasting adverse impact on objective, device-measured physical activity, particularly among Asian individuals. Successful management of patients with an ICD necessitates specific attention to clinically relevant behavioral outcomes, aiming to expedite recovery and facilitate a return to activities of daily living, with a particular focus on the Asian population.

An Algorithm for Strapdown Airborne Gravity Disturbance Vector Measurement Based on High-Precision Navigation and EGM2008.

Attitude errors, accelerometer bias, the gravity disturbance vector, and their coupling are the primary factors obstructing strapdown airborne vector gravimetry. This paper takes the geocentric inertial frame as a reference and solves the kinematic equations of its motion and its errors of the body frame and local geographic frame in the Lie group, respectively; the attitude accuracy is improved through a high-precision navigation algorithm. The constant accelerometer bias is estimated through Kalman filtering and is deducted from the accelerometer output to eliminate its influence. Based on the EGM2008 model, the low-frequency components of the gravity disturbance vector are corrected. The gravity disturbance vectors after model data fusion were low-pass filtered to obtain the ultimate results. This method was applied to flight experimental data in the South China Sea, and a gravity anomaly accuracy of better than 0.5 mGal, a northward gravity disturbance accuracy of 0.85 mGal, and an eastward gravity disturbance accuracy of 4.0 mGal were obtained, with a spatial resolution of approximately 4.8 km.

Development of a reliable adaptive estimation approach for a low‐cost attitude and heading reference system

AbstractA main challenge within inertial navigation systems involves attitude determination, which refers to the process of estimating the angles that define orientations. A new algorithm is presented to estimate a reliable and proper model for a low‐cost attitude and heading reference system (AHRS), while also, including long‐term global navigation satellite system (GNSS) outages with various dynamical manoeuvres. In the proposed approach, the initial AHRS model is continuously refined at each time step through the incorporation of complementary terms using a new prediction method. These complementary terms are determined using attitude information from the accelerometers output and the GNSS as reference systems. The proposed estimation technique undergoes mathematical analysis to ascertain stochastic stability and is further assessed through real‐world aerial experimental tests conducted with hardware‐in‐the‐loop mechanisation. The obtained results demonstrate a significant enhancement in the reliability and accuracy of the AHRS through the proposed estimation algorithm, even under GNSS blockages. The comparative results highlight the higher performance of the suggested data fusion scheme in providing a reliable and accurate model for the AHRS in normal conditions and assisting a powerful neural network‐based algorithm to provide reliable heading information even during long‐term GNSS outages under dynamical manoeuvres.

A Temperature Compensation Approach for Micro-Electro-Mechanical Systems Accelerometer Based on Gated Recurrent Unit-Attention and Robust Local Mean Decomposition-Sample Entropy-Time-Frequency Peak Filtering.

MEMS accelerometers are significantly impacted by temperature and noise, leading to a considerable compromise in their accuracy. In response to this challenge, we propose a parallel denoising and temperature compensation fusion algorithm for MEMS accelerometers based on RLMD-SE-TFPF and GRU-attention. Firstly, we utilize robust local mean decomposition (RLMD) to decompose the output signal of the accelerometer into a series of product function (PF) signals and a residual signal. Secondly, we employ sample entropy (SE) to classify the decomposed signals, categorizing them into noise segments, mixed segments, and temperature drift segments. Next, we utilize the time-frequency peak filtering (TFPF) algorithm with varying window lengths to separately denoise the noise and mixed signal segments, enabling subsequent signal reconstruction and training. Considering the strong inertia of the temperature signal, we innovatively introduce the accelerometer's output time series as the model input when training the temperature compensation model. We incorporate gated recurrent unit (GRU) and attention modules, proposing a novel GRU-MLP-attention model (GMAN) architecture. Simulation experiments demonstrate the effectiveness of our proposed fusion algorithm. After processing the accelerometer output signal through the RLMD-SE-TFPF denoising algorithm and the GMAN temperature drift compensation model, the acceleration random walk is reduced by 96.11%, with values of 0.23032 g/h/Hz for the original accelerometer output signal and 0.00895695 g/h/Hz for the processed signal.

Deep learning of movement behavior profiles and their association with markers of cardiometabolic health

BackgroundTraditionally, existing studies assessing the health associations of accelerometer-measured movement behaviors have been performed with few averaged values, mainly representing the duration of physical activities and sedentary behaviors. Such averaged values cannot naturally capture the complex interplay between the duration, timing, and patterns of accumulation of movement behaviors, that altogether may be codependently related to health outcomes in adults. In this study, we introduce a novel approach to visually represent recorded movement behaviors as images using original accelerometer outputs. Subsequently, we utilize these images for cluster analysis employing deep convolutional autoencoders.MethodsOur method involves converting minute-by-minute accelerometer outputs (activity counts) into a 2D image format, capturing the entire spectrum of movement behaviors performed by each participant. By utilizing convolutional autoencoders, we enable the learning of these image-based representations. Subsequently, we apply the K-means algorithm to cluster these learned representations. We used data from 1812 adult (20–65 years) participants in the National Health and Nutrition Examination Survey (NHANES, 2003–2006 cycles) study who worn a hip-worn accelerometer for 7 seven consecutive days and provided valid accelerometer data.ResultsDeep convolutional autoencoders were able to learn the image representation, encompassing the entire spectrum of movement behaviors. The images were encoded into 32 latent variables, and cluster analysis based on these learned representations for the movement behavior images resulted in the identification of four distinct movement behavior profiles characterized by varying levels, timing, and patterns of accumulation of movement behaviors. After adjusting for potential covariates, the movement behavior profile characterized as “Early-morning movers” and the profile characterized as “Highest activity” both had lower levels of insulin (P < 0.01 for both), triglycerides (P < 0.05 and P < 0.01, respectively), HOMA-IR (P < 0.01 for both), and plasma glucose (P < 0.05 and P < 0.1, respectively) compared to the “Lowest activity” profile. No significant differences were observed for the “Least sedentary movers” profile compared to the “Lowest activity” profile.ConclusionsDeep learning of movement behavior profiles revealed that, in addition to duration and patterns of movement behaviors, the timing of physical activity may also be crucial for gaining additional health benefits.

Influence of keel impacts and laying hen behavior on keel bone damage

Keel bone damage, which presents as fractures and/or deviations of the keel, has been detected in laying hens housed in all types of systems. Factors leading to keel bone damage in hens housed with limited vertical space, such as those housed in furnished systems, are not well understood, and are the topic of this study. Ten focal hens from each of 12 furnished cages (4 rooms of 3 cages) were fitted with keel mounted tri-axial accelerometers. Their behavior was video recorded continuously over two 3-wk trials: the first when the hens were between 52 and 60 wk of age, and the second approximately 20 wk later. The integrity of each hen's keel was evaluated at the start and end of each 3-wk trial using digital computed tomography. We identified predominant behaviors associated with acceleration events sustained at the keel (collisions, aggressive interactions and grooming) by pairing accelerometer outputs with video data. For each recorded acceleration event we calculated the acceleration magnitudes as the maximum summed acceleration recorded during the event, and by calculating the area under the acceleration curve. A principle components analysis, which was used as a data reduction technique, resulted in the identification of 4 components that were used in a subsequent regression analysis. A key finding is that the number of collisions a hen has with structures in her environment, and the number of aggressive interactions that a hen is involved, each affect the likelihood that she will develop 1 or more fractures within a 3-wk time span. This relationship between hen behavior and keel fracture formation was independent of the magnitude of acceleration involved in the event. Observed behavior did not have an impact on the formation of keel bone deviations, further supporting reports that the mechanisms underlying the 2 types of keel bone damage are different.

Activity Recognition Using Different Sensor Modalities and Deep Learning

In recent years, human activity monitoring and recognition have gained importance in providing valuable information to improve the quality of life. A lack of activity can cause health problems including falling, depression, and decreased mobility. Continuous activity monitoring can be useful to prevent progressive health problems. With this purpose, this study presents a wireless smart insole with four force-sensitive resistors (FSRs) that monitor foot contact states during activities for both indoor and outdoor use. The designed insole is a compact solution and provides walking comfort with a slim and flexible structure. Moreover, the inertial measurement unit (IMU) sensors designed in our previous study were used to collect 3-axis accelerometer and 3-axis gyroscope outputs. Smart insoles were located in the shoe sole for both right and left feet, and two IMU sensors were attached to the thigh area of each leg. The sensor outputs were collected and recorded from forty healthy volunteers for eight different gait-based activities including walking uphill and descending stairs. The obtained datasets were separated into three categories; foot contact states, the combination of acceleration and gyroscope outputs, and a set of all sensor outputs. The dataset for each category was separately fed into deep learning algorithms, namely, convolutional long–short-term memory neural networks. The performance of each neural network for each category type was examined. The results show that the neural network using only foot contact states presents 90.1% accuracy and provides better performance than the combination of acceleration and gyroscope datasets for activity recognition. Moreover, the neural network presents the best results with 93.4% accuracy using a combination of all the data compared with the other two categories.

Note on Coarse Alignment of Gyro-Free Inertial Navigation System

In this note, the feasibility of initial alignment of a gyro-free inertial navigation system (GF-INS) is investigated. Initial roll and initial pitch are obtained using leveling of conventional INS since centripetal acceleration is very small. The equation for the initial heading cannot be used since the GF inertial measurement unit (IMU) cannot directly measure the Earth rate. A new equation is derived to obtain the initial heading from GF-IMU accelerometer outputs. Initial heading is expressed in the accelerometer outputs of two configurations, which satisfies a specific condition among 15 GF-IMU configurations presented in the literature. The initial heading error to arrangement and accelerometer error is quantitatively analyzed from the initial heading calculation equation of GF-INS and the initial heading error analysis of the general INS. The initial heading error is investigated when gyroscopes are used with GF-IMU. The results show that the initial heading error depends more on the performance of the gyroscope than that of the accelerometer, and the initial heading cannot be obtained within a practical error level by using only GF-IMU, even when an extremely accurate accelerometer is used. Therefore, aiding sensors have to be used in order to have a practical initial heading.

Alignment scheme optimization when gyro accuracy inconsistency for inertial navigation system

Rotation modulation technology can average out constant errors in fiber optic gyro (FOG) rotational inertial navigation system (RINS). As a result, angle random walk (ARW) of FOGs plays a significant role in the precision of initial alignment. Using different gyros as horizontal gyros gradually transitioned from theory to practical application with the advancement of design philosophy. The theoretical formula between azimuth alignment accuracy and random walk coefficients (RWCs) of two horizontal gyros is deduced in this paper. A completely new multiple rotations and stationary alignment scheme is suggested. A recursive least-squares filter with azimuth variation in body reference frame and accelerometer outputs in navigation frame as measurements is constructed. The efficiency of this method is demonstrated through simulation and experimentation. The findings demonstrate that using this formula and alignment scheme, azimuth alignment accuracy could be greatly increased, which can be improved by 10.85% compared to common continuous rotation alignment scheme.

High performance lunar constellation for navigation services to Moon orbiting users

In order to assist the ever growing number of missions planned to visit the Moon, an international effort is being made to design and commission a communication and navigation lunar constellation of satellites. To this purpose, the goal of this study is to provide first a strategy to help the design of a small-sized lunar constellation, entailing different objectives and users to assist. Then, various optimised constellations are tested including the orbital users’ filtering architectures, extracting insights on the state estimation achievable and its capability to support critical GNC tasks.Given the early stage design phase in which this study is carried out, the constellation design does not entail strong and clearly defined requirements yet. For such reason, a multi-objective optimisation (MOO) has been employed, which proved to be extremely beneficial, ensuring a correct amount of versatility to provide a large set of Pareto-optimal solutions. Different objectives associated to both various services and users have been put in place, exploring a search space based on classical Keplerian elements of the constellation servicers.To compare the resulting constellations and validate correctly the performance from a navigation perspective the users navigation scheme has been considered to be based on a GNSS/INS tightly coupled formulation. The GNSS receiver measurements are fused with accelerometer outputs through a navigation filter, providing thus a strategy where the absence of GNSS observables is partially covered by an inertial propagation of the on-board state. Attention is here focused on polar orbiters operational in Low Lunar Orbit (LLO) environment and the obtained results showed navigation capabilities good enough to enable correct execution of missions in the low lunar orbital region.

Assessing Physical Activity Levels among Chinese College Students by BMI, HR, and Multi-Sensor Activity Monitors.

In a laboratory experiment, 100 college students, 18-25 years old, wore the SenseWear Pro3 Armband™ (SWA; BodyMedia, Inc., Pittsburg, PA, USA) and performed 7 different physical activities. EE was measured by indirect calorimetry, while body motion and accelerations were measured with an SWA accelerometer. Special attention was paid to the analysis of unidirectional and three-directional accelerometer output. Seven physical activities were recorded and distinguished by SWA, and different physical activities demonstrated different data features. The mean values of acceleration ACz (longitudinal accel point, axis Z) and VM (vector magnitude) were significantly different (p = 0.000, p < 0.05) for different physical activities, whereas no significant difference was found in one single physical activity with varied speeds (p = 0.9486, p > 0.05). When all physical activities were included in a correlation regression analysis, a strong linear correlation between the EE and accelerometer reporting value was found. According to the correlation analysis, sex, BMI, HR, ACz, and VM were independent variables, and the EE algorithm model demonstrated a high correlation coefficient R2 value of 0.7. The predictive energy consumption model of physical activity based on multi-sensor physical activity monitors, BMI, and HR demonstrated high accuracy and can be applied to daily physical activity monitoring among Chinese collegiate students.

Measurement of Physical Activity Using Accelerometry in Persons With Multiple Sclerosis

The consequences of multiple sclerosis (MS), particularly gait and walking dysfunction, may obfuscate (i.e., make unclear in meaning) the measurement of physical activity using body-worn motion sensors, notably accelerometers. This paper is based on an invited keynote lecture given at the 8th International Conference on Ambulatory Monitoring of Physical Activity and Movement, June 2022, and provides an overview of studies applying accelerometers for the measurement of physical activity behavior in MS. The overview includes initial research uncovering a conundrum with the interpretation of activity counts from accelerometers as a measure of physical activity. It then reviews research on calibration of accelerometer output based on its association with energy expenditure in yielding a biologically based metric for studying physical activity in MS. The paper concludes with other applications and lessons learned for guiding future research on physical activity measurement using accelerometry in MS and other populations with neurological diseases and conditions.

Improving measurement performance via fusion of classical and quantum accelerometers

AbstractWhile quantum accelerometers sense with extremely low drift and low bias, their practical sensing capabilities face at least two limitations compared with classical accelerometers: a lower sample rate due to cold atom interrogation time; and a reduced dynamic range due to signal phase wrapping. In this paper, we propose a maximum likelihood probabilistic data fusion method, under which the actual phase of the quantum accelerometer can be unwrapped by fusing it with the output of a classical accelerometer on the platform. Consequently, the recovered measurement from the quantum accelerometer is used to estimate bias and drift of the classical accelerometer which is then removed from the system output. We demonstrate the enhanced error performance achieved by the proposed fusion method using a simulated 1D accelerometer precision test scenario. We conclude with a discussion on fusion error and potential solutions.

Joint regression modelling of intensity and timing of accelerometer counts.

Accelerometers are commonly used in human medical and public health research to measure physical movement, which is relevant in a wide range of studies, from physical activity and sleep behaviours studies, to identification of movement patterns in people affected by diseases of the locomotor system and prediction of risk of injury in high performance sports. The accelerometer output provides the intensity (activity count) and timing (timestamp) of the movement, which can be used to define bouts of activity (periods of sustained movement of a given intensity). In some contexts, it may be important to include both dimensions to obtain a broader and deeper understanding of the phenomenon under study. Such is the case of a large-scale epidemiological investigation on the daily and weekly physical activity behaviours of school-aged children enrolled in the UK Millennium Cohort Study, which has motivated the present article. I present a statistical approach to joint modelling of intensity and timing of activity bouts that takes advantage of the circular nature of the timing. The model, which accounts for the longitudinal structure of the observations, is remarkably simple to implement using standard statistical software.

Stable physical activity tracking during children's guided active play.

ObjectivePhysical activity (PA) decreases during childhood with PA tracking statistics showing moderate coefficients in early childhood moving to poor coefficients in late childhood. The age-related instability of PA tracking is attributed to variations in age-related PA behaviors when quantifying PA in different settings (in/out of school, sports camps, and habitual PA). This study has examined the stability of age-related PA for children (from 7 to 11 years) during self-paced guided active play (GAP) sessions separated by 12 months.MethodsChildren (n = 65) recruited from community camps in two consecutive years were assessed for growth and PA participation during GAP sessions (1 h.d−1 on 2 d.wk−1) using cooperative games. Accelerometer outputs were used to quantify PA and estimate energy expenditure (EE) and moderate-to-vigorous PA (MVPA). Tracking statistics were assessed by Spearman coefficients (r) and agreement scores for percentile rankings (Kappa (k)) after a 1-year interval.ResultsGrouped data for PA tracking (EE) over the 1-year interval showed strong coefficients (r = 0.88, p = 0.001) and moderate agreement scores (k = 0.54). Boys and girls showed similar results. During the 1-year interval, the MVPA tracking coefficient was r = 0.54 (p = 0.01) with a moderate k score of 0.47. Age-related PA (EE) tracking coefficients at 7 (r = 0.90), 8 (r = 0.88), 10 (r = 0.82), and 11 (r = 0.83) years showed strong coefficients except at 9 years (r = 0.51).ConclusionChildren's age-related PA tracking (EE and MVPA) is stable during self-paced GAP assessed over a 1-year interval. Since PA tracking is stable in self-paced GAP, GAP should be included in children's PA intervention strategies to improve health and fitness.

Soil mass movement monitoring for landslide detection using low-cost accelerometer sensor as inclinometer

This paper presents soil mass movement monitoring for landslides detection using low-cost MEMS accelerometer as inclinometer. Commercial inclinometers for geotechnical ground observations are quite expensive. This research aims to study and develop low-cost inclinometer as an alternative using accelerometer. The output of the low-cost accelerometer is noisy and fluctuated make it not suitable for accurate measurement device. We solved this problem in this paper using moving average filter. The digital filter algorithm was tested and showed promising results.

Wavelet DE-noising and Kalman Filtering of MEMS Sensors for Autonomous Latitude Determination

Background: There is a task of autonomous determination of the position latitude of unmoved vehicles. Also, there is another task of the initial value latitude determination as a prepared operation of gimbaled and strap-down inertial navigation systems. For both cases, it is necessary to have an inertial measurement unit (IMU) with triad gyroscopes and triad accelerometers. The output signals of micromechanical gyroscopes and accelerometers have large noise components when using the IMU in the micromachined electromechanical systems (MEMS) technology. Objective: Normally to filter output signals of MEMS sensors, averaging and filtering are used. However, for Kalman filtering, it is necessary to find the exact mathematical model of the sensors and a lot of their initial random characteristics. The study of the possibility of the wavelet transform usage to filter the output signals MEMS accelerometers and gyroscopes for the latitude autonomous determination was conducted in the paper. Method: The wavelet transform approach was used to filter the output signals of MEMS accelerometers and gyroscopes in order to improve the accuracy of the autonomous position latitude determination. The autonomous latitude determination efficiency of IMU based on MEMS gyroscopes and accelerometers has been experimentally confirmed. The projections of the Earth’s angular rate and gravitational acceleration were obtained from the MEMS IMU. After that, the signals of the IMU gyroscopes and accelerometers were filtered, using the wavelet ‘Daubechies’ in decomposition, and averaged. These signals were used in a computational algorithm to determine the latitude. Results: The results showed that Unlike the well-known Kalman filter wavelet de-noising reduced calculation error by almost twice. Conclusion: Wavelet de-noising could be used for output signals filtering of micromechanical gyroscopes and accelerometers for the autonomous determination of position latitude.

A Hybrid Algorithm for Noise Suppression of MEMS Accelerometer Based on the Improved VMD and TFPF

High-G MEMS accelerometer (HGMA) is a new type of sensor; it has been widely used in high precision measurement and control fields. Inevitably, the accelerometer output signal contains random noise caused by the accelerometer itself, the hardware circuit and other aspects. In order to denoise the HGMA’s output signal to improve the measurement accuracy, the improved VMD and TFPF hybrid denoising algorithm is proposed, which combines variational modal decomposition (VMD) and time-frequency peak filtering (TFPF). Firstly, VMD was optimized by the multi-objective particle swarm optimization (MOPSO), then the best decomposition parameters [kbest, abest] could be obtained, in which the permutation entropy (PE) and fuzzy entropy (FE) were selected for MOPSO as fitness functions. Secondly, the accelerometer voltage output signals were decomposed by the improved VMD, then some intrinsic mode functions (IMFs) were achieved. Thirdly, sample entropy (SE) was introduced to classify those IMFs into information-dominated IMFs or noise-dominated IMFs. Then, the short-window TFPF was selected for denoising information-dominated IMFs, while the long-window TFPF was selected for denoising noise-dominated IMFs, which can make denoising more targeted. After reconstruction, we obtained the accelerometer denoising signal. The denoising results of different denoising algorithms in the time and frequency domains were compared, and SNR and RMSE were taken as denoising indicators. The improved VMD and TFPF denoising method has a smaller signal distortion and stronger denoising ability, so it can be adopted to denoise the output signal of the High-G MEMS accelerometer to improve its accuracy.

A Novel Self-Temperature Compensation Method for Mode-Localized Accelerometers.

Mode-localized sensing paradigms applied to accelerometers have recently become popular research subjects. However, the output of mode-localized accelerometers is influenced by environment temperature due to the difference in the thermal properties of the coupling resonators and the temperature dependence of coupling stiffness. To improve the performance of mode-localized accelerometers against temperature, we proposed an in situ self-temperature compensation method by utilizing the resonant frequency besides of amplitude ratios, which can be implied online. Experimental results showed that there were nearly 79-times and 87-times improvement in zeros bias and scale factor, respectively.

Accelerometer-based estimation of oxygen uptake in adults with Down syndrome: vector magnitude vs. vertical axis.

Triaxial accelerometer output [vector magnitude (VM) counts] may better estimate physical activity intensity as reflected in the rate of oxygen uptake (V̇O2 ) than the traditional vertical axis (VA) counts in adults with Down syndrome (DS). This study examined the accuracy of VM vs. VA counts in estimating V̇O2 in adults with and without DS across different physical activities and sedentary behaviours. Sixteen adults with DS (10 men and 6 women; 31±15years) and 19 adults without DS (10 men and 9 women; 24±5years) performed 12 tasks. V̇O2 was measured by portable spirometer (K4b2 , Cosmed) and VM and VA with an accelerometer (wGT3X-BT, Actigraph). Vector magnitude and VA were significant predictors of V̇O2 in adults with DS (P<0.001; R2 =0.74 and 0.65, respectively) and adults without DS (P<0.001; P<0.001; R2 =0.75 and 0.61, respectively). Absolute error of prediction was significantly smaller for VM than VA for sitting, playing app, drawing, sweeping, standing and basketball (P≤0.005), but smaller for VA than VM for walking at 0.8m·s-1 (P=0.005). Bland-Altman plots for adults with and without DS indicated narrower limits of agreement for VM than VA (-5.57 to 5.57 and -6.44 to 6.44mL·kg-1 ·min-1 ; -6.21 to 6.17 and -7.75 to 7.74mL·kg-1 ·min-1 , respectively). Vector magnitude and VA are significant predictors of V̇O2 in adults with and without DS, yet VM more accurately estimated V̇O2 than VA for most tasks. Development of accelerometer-based prediction of physical activity levels in adults with and without DS may improve by utilising VM counts.