Multi-Dimensional and Objective Assessment of Motion Sickness Susceptibility Based on Machine Learning

Area of Science:

• Physiology
• Biomedical Engineering
• Machine Learning

Background:

• Motion sickness (MS) is increasingly prevalent due to evolving transportation and recreational activities.
• Accurate MS susceptibility assessment aids in preventing exposure and ensuring public health and task safety.

Purpose of the Study:

• Develop an objective, multi-dimensional model for assessing motion sickness susceptibility.
• Utilize physiological indicators to objectively reflect MS severity.
• Provide a reference for enhancing current MS assessment methods.

Main Methods:

• Induced motion sickness in 51 participants using Coriolis acceleration stimulation.
• Digitized clinical manifestations of MS with portable equipment.
• Developed and compared machine learning (ML) models based on objective parameters and physiological data.

Main Results:

• Identified gastric electrical activity, facial skin color, skin temperature, and nystagmus as indicators related to MS severity.
• The support vector machine (SVM) ML model achieved 88.24% accuracy, 91.43% sensitivity, and 81.25% specificity.
• Demonstrated the efficacy of ML in creating objective MS susceptibility assessment models.

Conclusions:

• Objective quantification of motion sickness severity is achievable through specific physiological indicators.
• Multi-dimensional, objective MS susceptibility assessment models based on machine learning are feasible and effective.