This study focuses on the use of pressure mapping technology to monitor individuals at risk of pressure ulcers during extended lying postures. While this technology is effective in evaluating pressure ulcer risk, it does not currently provide information about posture or mobility. Therefore, the researchers aimed to develop an automated approach that can detect different static lying postures and transitions between them. Nineteen healthy subjects participated in the study and adopted various sagittal and lateral lying postures. The researchers continuously monitored parameters related to interactions at the support surface and body movements. They then examined the derivative of each signal to identify transitions between postures. To predict a range of static postures, three machine learning algorithms (Naïve-Bayes, k-Nearest Neighbors, and Support Vector Machine classifiers) were assessed using a training model consisting of data from nine subjects. The accuracy of these algorithms was then validated using new input from ten test subjects. The results showed that the derivative signals effectively detected transitions between postures, with actimetry providing the most distinct signal perturbations. The machine learning algorithms achieved high accuracy in predicting the range of postures from new test data, ranging from 82% to 100% for Naïve-Bayes, 70% to 98% for k-Nearest Neighbors, and 69% to 100% for Support Vector Machine classifiers. Overall, this study demonstrates that combining machine learning algorithms with robust parameters from two monitoring systems allows for reliable detection of both static postures and their corresponding transitions which can support personalized strategies for preventing pressure ulcers.
Pressure mapping technology is a way to check if someone might get sores from sitting or lying down for too long, but it doesn't tell us how they are sitting or moving. Researchers wanted to find a way to automatically see different ways people lie down and change positions. They studied 19 healthy people who lay in different positions and watched how their bodies moved on the surface they were lying on. They found that certain signals could show when someone changed position, especially with the help of another monitoring system called actimetry. They tested three computer programs that could predict how someone was lying down, and they were pretty accurate. By using these programs and watching the signals from the monitoring systems, we can reliably tell what position someone is in and when they change positions."
Definitions
- Pressure mapping technology: A method used to check if someone might get sores from sitting or lying down for too long.
- Posture: The way someone sits or lies down.
- Mobility: How easily someone can move around.
- Sagittal: A word used to describe something happening from front to back.
- Lateral: A word used to describe something happening from side to side.
- Parameters: Measurements or factors that are important for understanding something.
- Interactions: When things affect each other or work together.
- Support surface: The thing that someone is sitting or lying on.
- Derivative signals: Special signs that show when something changes quickly.
- Actimetry: Another
Using Pressure Mapping Technology and Machine Learning to Monitor Individuals at Risk of Pressure Ulcers
Pressure ulcers are a serious medical condition that can cause significant pain, discomfort, and even death. They occur when the skin is exposed to prolonged pressure or friction, which can lead to tissue damage and infection. As such, it is important for healthcare professionals to monitor individuals who are at risk of developing pressure ulcers in order to prevent them from occurring.
In recent years, researchers have been exploring the use of pressure mapping technology as an effective way of evaluating an individual’s risk for developing a pressure ulcer. However, this technology does not provide information about posture or mobility. Therefore, researchers set out to develop an automated approach that could detect different static lying postures and transitions between them in order to better assess an individual’s risk of developing a pressure ulcer.
Study Design
The study included nineteen healthy subjects who adopted various sagittal (lying on one's back) and lateral (lying on one's side) lying postures while being continuously monitored by two systems: actimetry (measuring body movements) and support surface interaction parameters (measuring interactions with the support surface). The researchers then examined the derivative signals from each system in order to identify transitions between postures.
Three machine learning algorithms were used in order to predict a range of static postures: Naïve-Bayes classifier, k-Nearest Neighbors classifier, and Support Vector Machine classifiers. These algorithms were assessed using data from nine test subjects before being validated with new input from ten test subjects.
Results
The results showed that the derivative signals effectively detected transitions between postures; actimetry provided the most distinct signal perturbations compared with other parameters measured by both systems combined together. Additionally, all three machine learning algorithms achieved high accuracy in predicting the range of postures from new test data; Naïve-Bayes had accuracies ranging from 82% - 100%, k-Nearest Neighbors had accuracies ranging from 70% - 98%, while Support Vector Machines had accuracies ranging from 69% - 100%.
Conclusion
Overall, this study demonstrates that combining machine learning algorithms with robust parameters obtained through monitoring systems allows for reliable detection of both static postures and their corresponding transitions which can support personalized strategies for preventing pressure ulcers among individuals at risk.
Background: Pressure mapping technology has been adapted to monitor over prolonged periods to evaluate pressure ulcer risk in individuals during extended lying postures. However, temporal pressure distribution signals are not currently used to identify posture or mobility. The present study was designed to examine the potential of an automated approach for the detection of a range of static lying postures and corresponding transitions between postures.Methods: Healthy subjects (n = 19) adopted a range of sagittal and lateral lying postures. Parameters reflecting both the interactions at the support surface and body movements were continuously monitored. Subsequently, the derivative of each signal was examined to identify transitions between postures. Three machine learning algorithms, namely Na{\"i}ve-Bayes, k-Nearest Neighbors and Support Vector Machine classifiers, were assessed to predict a range of static postures, established with a training model (n = 9) and validated with new input from test data (n = 10).Findings: Results showed that the derivative signals provided a means to detect transitions between postures, with actimetry providing the most distinct signal perturbations. The accuracy in predicting the range of postures from new test data ranged between 82%--100%, 70%--98% and 69%--100% for Na{\"i}ve-Bayes, k-Nearest Neighbors and Support Vector Machine classifiers, respectively.Interpretation: The present study demonstrated that detection of both static postures and their corresponding transitions was achieved by combining machine learning algorithms with robust parameters from two monitoring systems. This approach has the potential to provide reliable indicators of posture and mobility, to support personalised pressure ulcer prevention strategies.
