The application of artificial intelligence in spine surgery

Abstract

Due to its obvious advantages in processing big data and image information, the combination of artificial intelligence and medical care may profoundly change medical practice and promote the gradual transition from traditional clinical care to precision medicine mode. In this artical, we reviewed the relevant literatures and found that artificial intelligence was widely used in spine surgery. The application scenarios included etiology, diagnosis, treatment, postoperative prognosis and decision support systems of spinal diseases. The shift to artificial intelligence model in medicine constantly improved the level of doctors' diagnosis and treatment and the development of orthopedics.

Current Trends in Artificial Intelligence-Assisted Spine Surgery: A Systematic Review

“…spinal research, preoperative planning, decision making, postoperative outcome predictions, intraoperative navigation-assisted surgery, and selection of patients for spinal surgery (Fig. 5) [5,[21][22][23][24][25][26].…”

“…Currently, the use of AI in spine surgery is in its early developmental phase, encountering several obstacles such as fragmented data, insufficient integration, and limited clinical implementation [ 5 ]. However, as AI is being applied across the entire medical field, AI-based research in challenging and high-risk spinal surgeries is ongoing.…”

Machine Learning in Spine Surgery: A Narrative Review

“…13,14 ML can aid in diagnostic and clinical decision-making by training on large amounts of images. 2 In addition, ML may reduce erroneous radiological findings because substantial portions of asymptomatic patients exhibit benign abnormalities of the spine 15 (Table 2).…”

“…Surgeons are continually incorporating ML into their practice. 2,10,12,13,122 A novel future direction is the development of mixed reality (MR). MR uses complex image overlay projected on the natural visual field.…”

“…Different mathematical approaches include linear regression, classification, clustering, and dimensionality reduction. 2 ML models are then optimized by comparing the accuracy of different patterns that are identified within a data set. In using the same example referenced previously, if a model identifies hypertension as a strong prognosticator of surgical site infection, future iterations will prioritize hypertension to refine predictions.…”