Machine Learning in Spine Oncology: A Narrative Review
Seth B. Wilson,Jacob Ward,Vikas Munjal,Chi Shing Adrian Lam,Mayur Patel,Ping Zhang,David S. Xu,Vikram B. Chakravarthy
DOI: https://doi.org/10.1177/21925682241261342
2024-06-13
Global Spine Journal
Abstract:Global Spine Journal, Ahead of Print. Study DesignNarrative Review.ObjectiveMachine learning (ML) is one of the latest advancements in artificial intelligence used in medicine and surgery with the potential to significantly impact the way physicians diagnose, prognose, and treat spine tumors. In the realm of spine oncology, ML is utilized to analyze and interpret medical imaging and classify tumors with incredible accuracy. The authors present a narrative review that specifically addresses the use of machine learning in spine oncology.MethodsThis study was conducted in accordance with the Preferred Reporting Items of Systematic Reviews and Meta-Analysis (PRISMA) methodology. A systematic review of the literature in the PubMed, EMBASE, Web of Science, Scopus, and Cochrane Library databases since inception was performed to present all clinical studies with the search terms '[[Machine Learning] OR [Artificial Intelligence]] AND [[Spine Oncology] OR [Spine Cancer]]'. Data included studies that were extracted and included algorithms, training and test size, outcomes reported. Studies were separated based on the type of tumor investigated using the machine learning algorithms into primary, metastatic, both, and intradural. A minimum of 2 independent reviewers conducted the study appraisal, data abstraction, and quality assessments of the studies.ResultsForty-five studies met inclusion criteria out of 480 references screened from the initial search results. Studies were grouped by metastatic, primary, and intradural tumors. The majority of ML studies relevant to spine oncology focused on utilizing a mixture of clinical and imaging features to risk stratify mortality and frailty. Overall, these studies showed that ML is a helpful tool in tumor detection, differentiation, segmentation, predicting survival, predicting readmission rates of patients with either primary, metastatic, or intradural spine tumors.ConclusionSpecialized neural networks and deep learning algorithms have shown to be highly effective at predicting malignant probability and aid in diagnosis. ML algorithms can predict the risk of tumor recurrence or progression based on imaging and clinical features. Additionally, ML can optimize treatment planning, such as predicting radiotherapy dose distribution to the tumor and surrounding normal tissue or in surgical resection planning. It has the potential to significantly enhance the accuracy and efficiency of health care delivery, leading to improved patient outcomes.
clinical neurology,orthopedics