Artificial Intelligence and Minimally Invasive Spine Surgery

Abstract

At present, many AI models are used in medical image analysis and disease detection. Coupled with extended understanding of imaging and clinical phenotypes toward a more precision-based approach for managing spine patients, the multidimensionality of data-driven results and analysis become exponentially more complex. However, with the use of AI, ML platforms and efforts of spine surgeons, computer-aided diagnosis and detection, decision support and prognosis prediction, and computer-aided surgical education and assessment models seem to have reduced complications and improved outcomes with MISS. Like all medical specialties, MISS is subjected to changes in clinical practices based on cutting-edge research. Quality control systems should be employed to ensure the algorithms continue to be relevant as data collection and clinical practices change over time. Recent studies clearly demonstrate the power of AI, ML in pre-, intra-, and postoperative phases of spinal surgeries. Future work requires the integration of AI and ML models to combine pre-, intra-, and postoperative algorithms into a single model where the best preoperative planning, intraoperative surgical intervention, and postoperative follow-up work with associated risk, financial cost, and considerations can be suggested to surgeons. Such algorithms not only can benefit spine surgeons in their decision-making but also facilitate the delivery of high-quality healthcare to low resources settings and facilitate personalized surgical and postsurgical care with MISS.