Akademik Veri Yönetim Sistemi
Expert Systems with Applications, cilt.238, 2024 (SCI-Expanded)
Serious games have emerged as promising tools for physiotherapy, offering an engaging way to motivate patients and improve physical functions. Stress, a common physiological response, can impede patient performance and hinder recovery in physiotherapy. However, there is limited research on stress recognition for children during physiotherapeutic serious games. This study proposes a deep and machine learning based method for stress detection from facial images of children. The contributions include the utilization of a novel dataset comprising videos of 25 children, including children with specific conditions, and the exploration of using adults’ facial images for data augmentation. The proposed method employs a modified deep network architecture, VGG-Face, for facial emotion recognition and utilizes three machine learning models for stress recognition. This study represents the first attempt to recognize stress in children's facial images during physiotherapy with serious games. The findings hold the potential to optimize patient outcomes and contribute to monitoring patients during home therapy.