Massey C, Latif U, Deer TR, Kalia H, Guirguis M, Bicket MC, Khatri N, Tolba R, Chakravarthy K, Budwany R, Lee DW, Patel AA, Moreira AM, Jimenez DA, Patel RG, Cowan I, Christiansen S, Wilson D, Chapman KB, Vu C, Manzi S, Nashi S, Waheed M, Shah A, Nasseri M, Buchanan P, Azeem N, Petersen EA, Schatman ME, Abd-Elsayed A, Sayed D, Diwan S
Abstract
Introduction: Artificial intelligence (AI) is rapidly evolving and becoming more ubiquitous. Significant advancements have been made in the last few years, driving rapidly increasing adoption. The scale of publications on AI makes it difficult to keep abreast of relevant findings. Objective: The ASPN Artificial Intelligence Guidelines are designed to help clinicians understand AI and implement it into their practice. This Neuron Project is designed to evolve with the changing landscape of AI. Methods: An expert panel was chosen to discuss and write the best practice guidelines on AI. The primary authors conducted a literature search on PubMed, cross-referencing key terms in pain management and AI. After a thorough review of the current literature, the information collected was divided into broad categories of potential benefits, potential harms, and ways to ensure the benefits outweigh the potential harms of AI. These guidelines include only the most essential aspects of AI that clinicians need to know and understand before implementing AI into their practice. Results: Over 12,000 articles were found using the above search results. Many articles were reviews and clinical guidelines. The framework created from these guidelines allowed authors to fill in knowledge gaps and discuss the most critical elements pain clinicians should comprehend about AI. Conclusion: All authors achieved consensus on guidelines for implementing and using AI in pain management following a process of critical review and edits by the entire group of authors. All authors approved final guidelines. Discussion: The field of artificial intelligence is rapidly growing. As it expands into healthcare, it is necessary to prevent breaches of sensitive data and potential harm to patients. Guidelines that constantly evolve and grow with expanding indications for AI are essential to maximize benefit and prevent harm. This paper is part of ASPN's Neuron Project and is designed to update continuously as this field evolves.
