The use of a machine learning approach to predict perceived stress and quality of life among caregivers of stroke patients

Background and aim of the work: Caregivers of stroke survivors play a crucial role in providing home care, which often involves significant stress and impacts their quality of life. Various factors, including caregiving responsibilities, work-life balance, and social support, influence caregivers’ well-being. This study aims to examine the quality of life and stress levels among caregivers of stroke survivors. Research design and Methods: A comprehensive survey was conducted using personalized questions and two validated instruments: the SF-36 Health Survey and the Perceived Stress Scale (PSS-10). Additionally, the SF-36 was employed for training an Artificial Intelligence (AI) model to predict perceived stress levels, generating estimated scores on the PSS-10. Results. The findings indicate that caregivers experience significant stress and have low quality of life scores. The AI model successfully predicted perceived stress levels, demonstrating the utility of combining health surveys with AI techniques for efficient stress assessment. Conclusions: Understanding the experiences and well-being of caregivers is essential for developing targeted interventions to support them. Improving caregivers’ quality of life can enhance the overall management of stroke-affected patients. (www.actabiomedica.it).