Personalization of convolutional neural networks within the stress detection task using heart rate variability data
Annotation
Stress detection is an active area of research with important implications for personal, occupational, and social health. Most modern approaches use features computed from multiple sensor modalities, i.e., grouping different types of data from multiple sources for processing. These include electrocardiogram, electrodermal activity, electromyogram, skin temperature, respiration, accelerometer data, etc. Also, traditional machine learning algorithms (decision tree, discriminant analysis, support vector machine, etc.) or fully-connected neural networks are mostly used. Using these methods requires large amounts of data. Researchers are considering different approaches to personalization or generalization of models relative to subjects, namely subject-independent and subject-dependent (initially personal or adapted) models. The aim of the presented work is to develop a method for detecting stress based on heart rate variability data, taking into account the process of personalization of neural networks. The use of a convolutional neural network is proposed. The dependence of accuracy on the length of the input signal is studied. The dependence of accuracy on the data dimensionality reduction layer (one-dimensional convolutional layer, maximizing and averaging pooling) used in the network is also considered. The importance of personalizing models is demonstrated to significantly increase the accuracy of models of specific subjects. It is shown that the proposed method, based on 60 intervals between heartbeats, makes it possible to binary determine whether a person is under stress. Personalization allowed increasing the accuracy from 91.8 % to 98.9 ± 2.6 %. The F1-score value increased from 0.907 to 0.983 ± 0.038. The proposed personalized networks can be used in systems for monitoring the functional state of a person. They can also be used as part of a system that grants or restricts access to private resources based on whether a person is currently at rest.
Keywords
Постоянный URL
Articles in current issue
- Modeling the illumination of the Earth’s surface to select the operating modes of the radiation source
- Luminescent dynamics of oxygen oxidation of Viburnum opulus L. in chitosan solutions with gold nanoparticles
- Dynamic surface control for omnidirectional mobile robot with full state constrains and input saturation
- Dual-wavelength digital holographic interferometry for technical applications
- Structural analysis of ZrO2 and TiO2 nanoparticles
- Investigation of polyvinyl butyral coatings with carbon quantum dots on the characteristics of silicon solar cells
- Numerical algorithm for finding the optimal composition of the reacting mixture on the basis of the reaction kinetic model
- Raman spectroscopy of nanocomposites ZnO/ZnS and ZnO/ZnSe obtained by solvothermal-microwave synthesis method
- Emotion analysis of social network data using cluster based probabilistic neural network with data parallelism
- Assessing the possibility of using the method of image decomposition based on topological features to reduce entropy during image compression
- Implementation of neural networks in the method of multilevel component circuits
- Fuzzy logic controller algorithm for placing files in a data storage system
- Using topological data analysis for building Bayesan neural networks
- Method of modeling viscoelastic properties of oriented polymer materials using multi-barrier theory
- Design of microstrip patch antenna using Fennec Fox optimization with SSRR metamaterial for terahertz applications
- Algorithm for promptly maintaining the temperature regime of power amplification units of the radar transmitting complex based on a thermal model
- Convective heat transfer and hydrodynamics of flow at the endwall around a turbine blade under the influence of a magnetic field
- Methods of contactless registration of information signals for the audit of information security of power supply systems and networks
- Parameter estimation of permanent magnet synchronous motor
- Internal memory data protecting problems of the Renesas microcontrollers