Opportunities and challenges of machine learning models for prediction and diagnosis of spondylolisthesis: a systematic review

Abstract

Applications of machine learning algorithms in healthcare domain gained immense popularity and attracted research communities in the last decade. Interdisciplinary collaboration leads to development of new models to investigate issues related to the spondylolisthesis (slippage of one vertebrae over another) with promising results and large potential. This paper summarises available machine learning models to detect and predict spondylolisthesis. It would be a valuable resource from modelling and application perspective. We extracted papers by systematic searching of databases: Scopus, PubMed, IEEE, Google Scholar, ResearchGate, Springer and Elsevier with preset inclusion-exclusion criteria. Articles were analysed as per title, abstract, and full-text review. Finally, opportunities and challenges in this area is discussed. We examined the specific models and frameworks employed, and the overall performance achieved according to the metrics used at each work under study. Our findings indicate that machine learning model can provide high accuracy and outperforms in existing image processing techniques.