Artificial Intelligence Algorithms and Applications in Physical Therapy for Patients with Low Back Pain: Review of the Past Five Years

Document Type : Review Article

Authors

1 Assistant professor for women's health

2 student

Abstract

Our concise review aimed to understand the potential of artificial intelligence (AI) algorithms and applications in physical therapy and rehabilitation, specifically focusing on patients suffering from low back pain (LBP). The review outlined both the positive and negative implications of AI and machine learning techniques for enhancing health outcomes and promoting healthier lifestyles.
Methods: Studies were sourced from the WOS, Scopus, PubMed, and IEEE Xplore databases. Two reviewers conducted title/abstract and full-text screening. Data were gathered on model type, input variables, predicted outcomes, and machine learning techniques.
Results: Articles published between 2019 and 2024, focusing specifically on LBP and AI, including 12 papers showcasing a range of AI applications, including the "self back app," a mobile application designed for self-monitoring, prediction, and self-guided management of low back pain patients, demonstrating noteworthy positive outcomes. Additionally, convolutional neural networks (CNNs), a computational method, were utilized to analyze medical images and identify specific weak back muscles, aiding physiotherapists in designing targeted interventions. AI interventions were also employed to show the suitability of acute LBP patients for particular treatments based on collected data.
Conclusion: AI is transforming physiotherapy, particularly in treating LBP. This shift is evident in studies on real-time exercise feedback via mobile apps, analysis of trunk movement, and machine learning-based classification in individuals with chronic LBP. Furthermore, medical expert systems leveraging Bayesian networks offer innovative and personalized management strategies, redefining the future of physiotherapy

Keywords

Main Subjects