Natural language processing: using artificial intelligence to understand human language in orthopedics

Abstract

Natural language processing (NLP) describes the broad ield of artiicial intelligence by which computers are trained to understand and generate human language. Within healthcare research, NLP is commonly used for variable extraction and classiication/cohort identiication tasks. While these tools are becoming increasingly popular and available as both opensource and commercial products, there is a paucity of the literature within the orthopedic space describing the key tasks within these powerful pipelines. Curation and navigation of the electronic medical record are becoming increasingly onerous, and it is important for physicians and other healthcare professionals to understand potential methods of harnessing this large data resource. The purpose of this study is to provide an overview of the tasks required to develop an NLP pipeline for orthopedic research and present recent examples of successful implementations.

Prevalence and Predictors of Concomitant Meniscal Surgery During Pediatric and Adolescent ACL Reconstruction: Analysis of 4729 Patients Over 20 Years at a Tertiary-Care Regional Children's Hospital

“…NLP has been used for medical research in a number of specialties, including orthopaedic surgery, for cohort identification and variable extraction tasks using clinical documentation. 6 , 33 - 35 , 40 , 45 These techniques have been shown to be more accurate than traditional approaches (eg, diagnostic or billing code–based searches) in identifying cohorts of interest. 38 Specifically, using NLP, we were able to identify 20% more ACL surgery cases compared with using Current Procedural Terminology billing codes alone.…”

A practical guide to the implementation of artificial intelligence in orthopaedic research—Part 2: A technical introduction

“…While computer vision, speech recognition, robotics and expert systems are broad subdomains of AI in their own right, the present discussion will be limited to the description of computational techniques suitable for clinical research in orthopaedics without prerequisites in engineering disciplines [ 66 ]. Orthopaedic researchers wishing to delve deeper into the technical workings of specific models and the interpretation of their outputs are referred to additional literature on the subject [ 16 , 33 , 40 , 49 , 50 , 58 , 59 ].…”

Artificial intelligence (AI) and large data registries: Understanding the advantages and limitations of contemporary data sets for use in AI research

“…Several capabilities of AI have emerged that may complement existing and future large registries [10][11][12][13][14]. There are several advantages to the use of large data registries as it pertains to applying AI statistical methods for research analyses.…”