Marco Altini

Estimating Energy Expenditure Using Body-Worn Accelerometers: A Comparison of Methods, Sensors Number and Positioning

Several methods to estimate energy expenditure (EE) using body-worn sensors exist; however, quantifications of the differences in estimation error are missing. In this paper, we compare three prevalent EE estimation methods and five body locations to provide a basis for selecting among methods, sensors number, and positioning. We considered 1) counts-based estimation methods, 2) activity-specific estimation methods using METs lookup, and 3) activity-specific estimation methods using accelerometer features. The latter two estimation methods utilize subsequent activity classification and EE estimation steps. Furthermore, we analyzed accelerometer sensors number and on-body positioning to derive optimal EE estimation results during various daily activities. To evaluate our approach, we implemented a study with 15 participants that wore five accelerometer sensors while performing a wide range of sedentary, household, lifestyle, and gym activities at different intensities. Indirect calorimetry was used in parallel to obtain EE reference data. Results show that activity-specific estimation methods using accelerometer features can outperform counts-based methods by 88% and activity-specific methods using METs lookup for active clusters by 23%. No differences were found between activity-specific methods using METs lookup and using accelerometer features for sedentary clusters. For activity-specific estimation methods using accelerometer features, differences in EE estimation error between the best combinations of each number of sensors (1 to 5), analyzed with repeated measures ANOVA, were not significant. Thus, we conclude that choosing the best performing single sensor does not reduce EE estimation accuracy compared to a five sensors system and can reliably be used. However, EE estimation errors can increase up to 80% if a nonoptimal sensor location is chosen.

Energy expenditure estimation using wearable sensors: a new methodology for activity-specific models

Accurate estimation of Energy Expenditure (EE) in ambulatory settings is a key element in determining the causal relation between aspects of human behavior related to physical activity and health. We present a new methodology for activity-specific EE algorithms. The proposed methodology models activity clusters using specific parameters that capture differences in EE within a cluster, and combines these models with Metabolic Equivalents (METs) derived from the compendium of physical activities. We designed a protocol consisting of a wide set of sedentary, household, lifestyle and gym activities, and developed a new activity-specific EE algorithm applying the proposed methodology. The algorithm uses accelerometer (ACC) and heart rate (HR) data acquired by a single monitoring device, together with anthropometric variables, to predict EE. Our model recognizes six clusters of activities independent of the subject in 52.6 hours of recordings from 19 participants. Increases in EE estimation accuracy ranged from 18 to 31% compared to state of the art single and multi-sensor activity-specific methods.

Correlation Between Electrode-Tissue Impedance and Motion Artifact in Biopotential Recordings

Ambulatory monitoring of the electrocardiogram (ECG) is a highly relevant topic in personal healthcare. A key technical challenge is overcoming artifacts from motion in order to produce ECG signals capable of being used in clinical diagnosis by a cardiologist. An electrode-tissue impedance is a signal of significant interest in reducing the motion artifact in ECG recordings on the go. A wireless system containing an ultralow-power analog front-end ECG signal acquisition, as well as the electrode-tissue impedance, is used in a validation study on multiple subjects. The goal of this paper is to study the correlation between motion artifacts and skin electrode impedance for a variety of motion types and electrodes. We have found that the correlation of the electrode-tissue impedance with the motion artifact is highly dependent on the electrode design the impedance signal (real, imaginary, absolute impedance), and artifact types (e.g., push or pull electrodes). With the chosen electrodes, we found that the highest correlation was obtained for local electrode artifacts (push, pull, electrode) followed by local skin (stretch, twist, skin) and global artifacts (walk, jog, jump). The results show that the electrode-tissue impedance can correlate with the motion artifacts for local disturbance of the electrodes and that the impedance signals can be used in motion artifact removal techniques such as adaptive filtering.

An ECG patch combining a customized ultra-low-power ECG SoC with Bluetooth low energy for long term ambulatory monitoring

This paper presents the development of an ECG patch aiming at long term patient monitoring. The use of the recently standardized Bluetooth Low Energy (BLE) technology, together with a customized ultra-low-power ECG System on Chip (ECG SoC). including Digital Signal Processing (DSP) capabilities, enables the design of ultra low power systems able to continuously monitor patients, performing on board signal processing such as filtering, data compression, beat detection and motion artifact removal along with all the advantages provided by a standard radio technology such as Bluetooth. Early tests show how combining the ECG SoC and BLE leads to a total current consumption of only 500μA at 3.7V, while computing beat detection and transmitting heart rate remotely via BLE. This allows up to one month lifetime with a 400mAh Li-Po battery.

Unsupervised activity clustering to estimate energy expenditure with a single body sensor

Body sensor networks (BSNs) have provided the opportunity to monitor energy expenditure (EE) in daily life and with that information help reduce sedentary behavior and ultimately improve human health. Current approaches for EE estimation using BSNs require tedious annotation of activity types and multiple body sensor nodes during data collection and high accuracy activity classifiers during post processing. These drawbacks impede deploying this technology in daily life — the primary motivation of using BSNs to monitor EE. With the goal of achieving the highest EE estimation accuracy with the least invasiveness and data collection effort, this paper presents an unsupervised, single-node solution for data collection and activity clustering. Motivated by a previous finding that clusters of similar activities tend to have similar regression models for estimating EE, we apply unsupervised clustering to implicitly group activities with homogeneous features and generate specific regression models for each activity cluster without requiring manual annotation. The framework therefore does not require specific activity classification, hence eliminating activity type labels. With leave-one-subject-out cross-validation across 10 subjects, an RMSE of 0.96 kcal/min was achieved, which is comparable to the activity-specific model and improves upon a single regression model.

An Android-based body area network gateway for mobile health applications

This paper presents a Body Area Network (BAN) gateway to Android mobile phones for mobile health applications. The proposed approach is based on a Secure Digital Input Output (SDIO) interface, which allows for long-term monitoring since the mobile phone hardware can be extended in order to operate with ultra low-power radios. The software architecture implemented on the mobile phone enables different features; data can be displayed, further processed or sent to a remote server exploiting the WLAN or 3G networks. Moreover, the system allows to configure thresholds on the measured parameters and to automatically send alerts such as SMS messages and emails based on these values. The system is illustrated for the case of ambulatory ECG monitoring.

Combining wearable accelerometer and physiological data for activity and energy expenditure estimation

Physical Activity (PA) is one of the most important determinants of health. Wearable sensors have great potential for accurate assessment of PA (activity type and Energy Expenditure (EE)) in daily life. In this paper we investigate the benefit of multiple physiological signals (Heart Rate (HR), respiration rate, Galvanic Skin Response (GSR), skin humidity) as well as accelerometer (ACC) data from two locations (wrist - combining ACC, GSR and skin humidity - and chest - combining ACC and HR) on PA type and EE estimation. We implemented single regression, activity recognition and activity-specific EE models on data collected from 16 subjects, while performing a set of PAs, grouped into six clusters (lying, sedentary, dynamic, walking, biking and running). Our results show that combining ACC and physiological signals improves performance for activity recognition (by 2 and 8% for the chest and wrist) and EE (by 36 - chest - and 35% - wrist - for single regression models, and by 18 - chest - and 46% - wrist - for activity-specific models). Physiological signals other than HR showed a coarser relation with level of physical exertion, resulting in being better predictors of activity cluster type and separation between inactivity and activity than EE, due to the weak correlation to EE within an activity cluster.

Self-calibration of walking speed estimations using smartphone sensors

Activity recognition for human behavior monitoring is an important research topic in the field of mHealth, especially for aspects of physical activity linked to fitness and disease progress, such as walking and walking speed. Sensors embedded into smartphones recently enabled new opportunities for non invasive activity and walking speed inference. In this paper, we propose a data fusion approach to the problem of physical activity recognition and walking speed estimation using smartphones. Our architecture combines different sensors to take into account practical issues arising in realistic settings, such as variability in phone location and orientation. Additionally, we introduce a novel automatic calibration methodology combining accelerometer and GPS data while walking in unconstrained settings, in order to reduce walking speed estimation error at the individual level. The proposed system was validated in 20 participants while performing sedentary, household, ambulatory and sport activities, in both indoor laboratory and outdoor self-paced settings. We show that by combining accelerometer and gyroscope data, smartphone location can be distinguished between the two most commonly used positions (bag and pocket), regardless of phone orientation (97 % f-score). Location-specific activity recognition models can significantly improve activity recognition performance (p = 0.0010 <; α), especially helping in distinguishing activities involving similar motion patterns (91 % f-score overall, improvements between 4% and 11 % for walking and biking activities). Our proposed method to personalize walking speed estimates, by automatically calibrating walking speed estimation models during a short self-paced walk, reduced walking speed estimation error by 8.8% on average (p = 0.0012 <; α).

Estimating Oxygen Uptake During Nonsteady-State Activities and Transitions Using Wearable Sensors

In this paper, we present a method to estimate oxygen uptake (VO2) during daily life activities and transitions between them. First, we automatically locate transitions between activities and periods of nonsteady-state VO2. Subsequently, we propose and compare activity-specific linear functions to model steady-state activities and transition-specific nonlinear functions to model nonsteady-state activities and transitions. We evaluate our approach in study data from 22 participants that wore a combined accelerometer and heart rate sensor while performing a wide range of activities (clustered into lying, sedentary, dynamic/household, walking, biking and running), including many transitions between intensities, thus resulting in nonsteady-state VO2. Indirect calorimetry was used in parallel to obtain VO2 reference. VO2 estimation error during transitions between sedentary, household and walking activities could be reduced by 16% on average using the proposed approach, compared to state of the art methods.

Personalizing energy expenditure estimation using physiological signals normalization during activities of daily living

In this paper we propose a generic approach to reduce inter-individual variability of different physiological signals (HR, GSR and respiration) by automatically estimating normalization parameters (e.g. baseline and range). The proposed normalization procedure does not require a dedicated personal calibration during system setup. On the other hand, normalization parameters are estimated at system runtime from sedentary and low intensity activities of daily living (ADLs), such as lying and walking. When combined with activity-specific energy expenditure (EE) models, our normalization procedure improved EE estimation by 15 to 33% in a study group of 18 participants, compared to state of the art activity-specific EE models combining accelerometer and non-normalized physiological signals.