Key result
Mean skin temperature increased linearly from a normal to a high-stress state in 80% of subjects, and a Probabilistic Neural Network achieved an 88.75% overall classification accuracy for four stress levels.
Why the study?
Does skin temperature variation accurately identify changes in stress levels induced by the Stroop color word test in healthy volunteers?
Cross-Sectional (n=60)
Does skin temperature variation accurately identify changes in stress levels induced by the Stroop color word test in healthy volunteers?
p-value: p=<0.0001
Skin temperature variation is a reliable physiological marker for identifying and classifying stress levels.
Should not yet alter stress assessment protocols; hypothesis-generating for skin temperature-based neural networks.
[Purpose] Stress is a common factor of several diseases. Stress can be reduced through appropriate stress management and relaxation methods. In this study, variation in skin temperature (ST) was investigated as a primary measure for identifying changes in stress levels. Our results should be helpful for the development of a stress measurement tool based on multimodal signals. [Subjects] Sixty healthy volunteers (30 females and 30 males) of three different races (Malay, Chinese, and Indian) with a mean age of 22.5±2.5 years participated in this study. [Methods] The Stroop color word test was used to design a data acquisition protocol of 12.36 min for this experiment. ST variation was measured continuously during the Stroop colour word test and statistical features were computed. Further, descriptive analysis and stress levels were classified using a Probabilistic Neural Network (PNN) to find the optimum features. [Results] Among the 60 subjects, the mean ST of 48 subjects (80%) rose linearly from the normal state to the high-stress state. In addition, Malay subjects were more sensitive to stress than other two races as measured by the mean skin temperature. A maximum mean classification rate of 88% was achieved for the four different stress levels on all the subjects using PNN. [Conclusion] Our investigation proves that the mean ST is a reliable measure for identifying stress level changes and may be useful for designing a multimodal stress measurement system.
No takes yet. Share an insight, caveat, or question.
Karthikeyan et al. (2012) conducted a cross-sectional in Healthy volunteers (n=60). Stroop colour word test vs. Relaxation (normal) state was evaluated on Classification accuracy of four stress levels using mean skin temperature via Probabilistic Neural Network (p=<0.0001). Mean skin temperature increased linearly from a normal to a high-stress state in 80% of subjects, and a Probabilistic Neural Network achieved an 88.75% overall classification accuracy for four stress levels.
Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context: