Related Experiment Video
Updated: Jun 18, 2026
06:49
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
Published on: December 11, 2015
8.9K
Energy Expenditure Prediction from Accelerometry Data Using Long Short-Term Memory Recurrent Neural Networks.
Martin Vibæk1, Abdolrahman Peimankar1, Uffe Kock Wiil1
1SDU Health Informatics and Technology, The Maersk Mc-Kinney Moller Institute, University of Southern Denmark, 5230 Odense, Denmark.
Sensors (Basel, Switzerland)
|April 27, 2024
Summary
New deep learning models accurately estimate energy expenditure using accelerometry data. Recurrent neural networks analyzing movement patterns show improved accuracy for physical activity monitoring in children.
Area of Science:
- Biomedical Engineering
- Wearable Technology
- Machine Learning in Health
Background:
- Accurate estimation of energy expenditure (EE) is crucial for physical activity (PA) interventions and population surveillance.
- Existing accelerometry methods often overlook the temporal dynamics of movement data.
- Novel approaches are needed to enhance the precision of EE prediction from objective measurements.
Purpose of the Study:
- To investigate the efficacy of recurrent neural networks (RNNs) in predicting energy expenditure by utilizing the temporal aspects of accelerometry data.
- To compare the performance of RNN-based models against traditional methods like Multiple Linear Regression (MLR).
Main Methods:
- Collected accelerometry data from 33 children during a standardized activity protocol in their natural environment.
- Measured acceleration at multiple body locations: hip, wrist, thigh, and back.
- Modeled energy expenditure using Multiple Linear Regression (MLR), stacked Long Short-Term Memory (LSTM) networks, and combined Convolutional Neural Networks (CNN) and LSTM.
Main Results:
- The combined LSTM-CNN model achieved the highest prediction accuracy, with a correlation of 0.883 and a Mean Absolute Percentage Error (MAPE) of 13.9%.
- LSTM networks demonstrated superior performance (correlation: 0.882, MAPE: 14.22%) compared to MLR (correlation: 0.76, MAPE: 19.9%).
- Prediction errors for vigorous PA intensities were significantly higher (p < 0.01) than for sedentary, light, and moderate intensities.
Conclusions:
- Incorporating temporal movement data through deep learning significantly enhances energy expenditure prediction accuracy.
- Recurrent neural network architectures, particularly combined CNN-LSTM, offer a promising advancement over conventional methods.
- Further research is required to address the heightened prediction error observed during vigorous physical activity.
Related Concept Videos
01:12Energy Stored in a Capacitor
When an archer pulls the string in a bow, he saves the work done in the form of elastic potential energy. When he releases the string, the potential energy is released as kinetic energy of the arrow. A capacitor works on the same principle in which the work done is saved as electric potential energy. The potential energy (UC) could be calculated by measuring the work done (W) to charge the capacitor.
01:25Neural Circuits
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
01:10Energy Stored in Capacitors
A parallel plate capacitor, when connected to a battery, develops a potential difference across its plates. This potential difference is key to the operation of the capacitor, as it determines how much electrical energy the capacitor can store.
By integrating the equation that relates voltage and current in a capacitor, one can derive an equation for the voltage across the capacitor at any given time. This equation is crucial in understanding and predicting the behavior of capacitors in...
By integrating the equation that relates voltage and current in a capacitor, one can derive an equation for the voltage across the capacitor at any given time. This equation is crucial in understanding and predicting the behavior of capacitors in...
01:17Energy and Power Signals
In an electrical system with a resistor, voltage and current signals facilitate the measurement of power and energy across the resistor. For a continuous-time signal, the total energy over a time interval is defined as the integral of the square of the signal's magnitude over that interval. Mathematically, this is expressed as:
01:23Storage
A schema is a mental framework that helps individuals organize and interpret information. Schemata, formed from previous experiences, influence how we process new information: how we encode it, the inferences we make, and how we retrieve it. For instance, a schema for what a typical classroom looks like might include desks, a teacher's desk, a whiteboard, and students in such an environment. This expectation helps us quickly understand and navigate new classrooms without needing to analyze each...
01:18Long-Term Memory
Long-term memory is a relatively permanent type of memory, capable of storing vast amounts of information over extended periods. Its storage capacity is generally considered unlimited.
Long-term memory can be categorized into two primary types: explicit and implicit memory. Explicit memory, also known as declarative memory, involves the conscious recollection of information that we deliberately try to remember, recall, and articulate. This type of memory encompasses specific facts, events, and...
Long-term memory can be categorized into two primary types: explicit and implicit memory. Explicit memory, also known as declarative memory, involves the conscious recollection of information that we deliberately try to remember, recall, and articulate. This type of memory encompasses specific facts, events, and...