Postural Data Analysis using AI-powered Classification Models

Nowadays, more and more people are working remotely or in professions that require them to sit for long periods of time. Unfortunately, spending too much time in a seated position can lead to a range of physical and mental health problems, such as musculoskeletal discomfort, headaches, and respiratory issues. These problems are often exacerbated by poor posture, which is common when sitting for extended periods of time. To address this issue, we have developed a system for classifying sitting postures using sensors and machine learning algorithms, achieving 100% of accuracy with a set of seven fiber Bragg grating sensors. We have further optimized the multisensor system by studying the optimal number of sensors and their positioning on the spine, achieving over 95% accuracy in classifying upright, kyphotic, and lordotic positions with as little as only two devices.