Low back pain (LBP) stands as a leading cause of disability globally, impacting countless lives and placing a significant burden on healthcare systems. Effectively diagnosing and treating LBP often requires a comprehensive and personalized strategy, incorporating various assessment tools, imaging techniques, and innovative technologies. The increasing volume of data generated in LBP management has paved the way for the integration of artificial intelligence (AI) and, specifically, diagnosis computer systems, also known as computer-aided diagnosis (CAD). These systems are designed to support clinicians in enhancing the accuracy and efficiency of LBP diagnosis and treatment planning.
A thorough review of existing research exploring the application of diagnosis computer in chronic LBP was conducted, examining studies indexed in PubMed, Scopus, and Web of Science. The research utilized keywords such as “Artificial Intelligence,” “Machine Learning,” and “Computer Aided Diagnosis” in conjunction with terms like “Low Back Pain” and “Lumbar Spine.” This extensive search initially yielded over 1500 articles. After a rigorous screening process, focusing on abstract and full-text reviews, 57 studies were identified as meeting the inclusion criteria for analysis.
The primary applications of diagnosis computer systems in LBP management fall into two main categories: classification and regression. Classification systems are employed to identify or categorize specific conditions related to LBP, aiding in differential diagnosis. Regression models, on the other hand, provide quantitative assessments, generating numerical outputs that evaluate the severity or progression of LBP-related measures. The most successful diagnosis computer systems have been developed for analyzing imaging data to detect degenerative changes in the spine. These systems have demonstrated impressive accuracy rates, often exceeding 80%. However, the utility of diagnosis computer extends beyond image analysis. Significant progress has been made in utilizing these tools to analyze clinical data, biomechanical factors, electrophysiological measurements, and functional imaging, offering a holistic approach to LBP assessment.
While current findings highlight the promising role of diagnosis computer in improving LBP care, further research is essential. Future studies should focus on refining these systems, validating their performance in diverse clinical settings, and exploring their integration into routine clinical workflows. The continued development and implementation of diagnosis computer technologies hold significant potential to revolutionize LBP diagnosis and ultimately enhance patient outcomes.
