Role of machine learning in management of degenerative spondylolisthesis: a systematic review

Background: Machine learning is a field of artificial intelligence that allows a computer system to learn through repetitive processes and improve with experience. Precise study of medical data benefits early disease recognition, patient care, and community services. Methods: The purpose of this systematic review was to assess the evidence for effectiveness of machine learning and artificial intelligence in the management of spondylolisthesis. A literature search of published and unpublished articles resulted in the retrieval of more than 1000 potential studies on the subject area. Eight were reviewed according to inclusion criteria. Results: Expert medical doctors examined the pelvis and lumbar spine shape and orientation to diagnose spondylolisthesis. However, some shape and orientation parameters were misleading and unclear. Therefore, automatic diagnosis methods (classification methods) have been proposed to help medical doctors. The most important parameter of classification was found to be the grade of spondylolisthesis. Conclusions: Although the proposed results may be misleading, the studies provided evidence to suggest that two-thirds of the patients with grade I spondylolisthesis were stable enough to tolerate decompression without fusion, but that one-third of the patients appeared to develop instability over time. This instability often led to reoperation for spinal fusion at the level of listhesis. It is possible to create a predictive machine learning algorithm that is calibrated and accurate to predict discharge placement.