A recurrent neural network architecture to model physical activity energy expenditure in older people

Abstract

Through the quantification of physical activity energy expenditure (PAEE), health care monitoring has the potential to stimulate vital and healthy ageing, inducing behavioural changes in older people and linking these to personal health gains. To be able to measure PAEE in a health care perspective, methods from wearable accelerometers have been developed, however, mainly targeted towards younger people. Since elderly subjects differ in energy requirements and range of physical activities, the current models may not be suitable for estimating PAEE among the elderly. Furthermore, currently available methods seem to be either simple but non-generalizable or require elaborate (manual) feature construction steps. Because past activities influence present PAEE, we propose a modeling approach known for its ability to model sequential data, the recurrent neural network (RNN). To train the RNN for an elderly population, we used the growing old together validation (GOTOV) dataset with 34 healthy participants of 60 years and older (mean 65 years old), performing 16 different activities. We used accelerometers placed on wrist and ankle, and measurements of energy counts by means of indirect calorimetry. After optimization, we propose an architecture consisting of an RNN with 3 GRU layers and a feedforward network combining both accelerometer and participant-level data. Our efforts included switching mean to standard deviation for down-sampling the input data and combining temporal and static data (person-specific details such as age, weight, BMI). The resulting architecture produces accurate PAEE estimations while decreasing training input and time by a factor of 10. Subsequently, compared to the state-of-the-art, it is capable to integrate longer activity data which lead to more accurate estimations of low intensity activities EE. It can thus be employed to investigate associations of PAEE with vitality parameters of older people related to metabolic and cognitive health and mental well-being.

IMU-Based Energy Expenditure Estimation for Various Walking Conditions Using a Hybrid CNN–LSTM Model

“…However, it requires expensive equipment and wearing a mask to measure the volume of oxygen and carbon dioxide; therefore, its use during activities of daily life is limited. To overcome these limitations, studies have been conducted to estimate EE using wearable sensors, such as inertial measurement units (IMU) or electromyography (EMG) sensors [ 12 , 13 , 14 , 15 , 16 , 17 , 18 ]. Such sensors are small, lightweight, and relatively low-cost; therefore, they can be easily applied to a variety of wearable devices and used in daily life.…”

“…An IMU comprising an accelerometer and a gyroscope can estimate the kinematics [ 19 , 20 , 21 ] and kinetics [ 22 , 23 , 24 ] of human motion, which have been found to be correlated with the EE [ 25 , 26 , 27 ]. Accordingly, several studies have developed algorithms to estimate EE based on IMU signals using linear regression or machine learning techniques [ 12 , 13 , 14 , 15 , 16 , 17 , 18 ]. Most algorithms extract the sum or features from IMU signals during a certain period for use as inputs to the model.…”

“…Zhu et al [ 17 ] developed a convolutional neural network (CNN) model that automatically extracted features for EE estimation from fixed-size sequences of accelerometer signals and heart rate data. Paraschiakos et al [ 18 ] developed a recurrent neural network (RNN) that estimates EE from the standard deviation of accelerometer data and trained and tested the model on elderly people.…”

Wearable Smart Jacket for Coal Miners Using IoT

“…Using this architecture, manual work is reduced, which means that a separate person is no longer required to monitor the patient continuously. It also increases data rate and allows for data retrieval at any point in time [3]. As a result, the patients' health records are tracked and saved on an ongoing basis.…”