On the analysis of saturated pressure to detect fatigue

Marcos Faundez-Zanuy,Josep Lopez-Xarbau,Moises Diaz,Manuel Garnacho-Castaño

DOI: https://doi.org/10.1007/978-3-031-45461-5_4

2024-10-22

Abstract:This paper examines the saturation of pressure signals during various handwriting tasks, including drawings, cursive text, capital words text, and signature, under different levels of fatigue. Experimental results demonstrate a significant rise in the proportion of saturated samples following strenuous exercise in tasks performed without resting wrist. The analysis of saturation highlights significant differences when comparing the results to the baseline situation and strenuous fatigue.

Signal Processing,Machine Learning

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to detect the impact of fatigue on writing tasks by analyzing the saturation of pressure signals. Specifically, the research aims to explore the changes in pressure signal saturation when performing writing tasks at different fatigue levels, especially when not resting the wrist.

### Research Background
Online handwritten data collection (such as using a digitizer) has wide applications in fields such as e - security and e - health. However, fatigue can affect an individual's motor coordination ability and fine - movement control ability, which in turn affects handwriting features such as letter shape, speed, and spacing. These changes can be used as potential indicators for detecting fatigue.

### Research Questions
1. **The relationship between pressure signal saturation and fatigue**: The researchers proposed a new feature - the pressure signal saturation ratio (i.e., the ratio of the pressure signal exceeding the saturation threshold) and explored the changes of this feature at different fatigue levels.

2. **Performance differences in specific tasks**: The researchers analyzed the performance of nine different writing tasks at different fatigue levels to determine which tasks are more sensitive to fatigue.

3. **Statistical significance analysis**: Through the Wilcoxon rank - sum test, the researchers evaluated the statistical significance of changes in pressure signal saturation at different fatigue levels.

### Main Findings
- **High - change tasks**: For tasks requiring higher fine - movement control (such as pentagon copying, house drawing, Archimedes' spiral, and concentric circles), fatigue can lead to a significant increase in pressure signal saturation, up to five times.

- **Low - change tasks**: For tasks such as signatures, capital letters, spring diagrams, and cursive texts, fatigue has a relatively small impact on pressure signal saturation.

- **Statistical significance**: Through statistical analysis, the researchers found that some tasks (such as concentric circles and Archimedes' spiral) showed significant changes in pressure signal saturation under extreme fatigue conditions.

### Conclusions
This study introduced the pressure signal saturation ratio as a new feature in handwriting analysis and demonstrated that this feature has high sensitivity in detecting tasks without wrist rest. Future research can further explore the potential of this feature in other applications, such as automatic classification and biometrics.

### Formula Summary
The calculation formula for the pressure signal saturation ratio is:

\[
\text{saturated} = \frac{1}{n} \sum_{i = 1}^{n}[\text{pressure\_signal}_i\geq\text{sat\_level}]
\]

where:
- \(n\) is the length of the input pressure signal vector,
- \(\text{pressure\_signal}_i\) is the pressure signal value of the \(i\) - th sample,
- \(\text{sat\_level}\) is the given saturation threshold (depending on the digitizer model).