Artificial Intelligence in Musculoskeletal Imaging: A Paradigm Shift

Abstract

Artificial intelligence is upending many of our assumptions about the ability of computers to detect and diagnose diseases on medical images. Deep learning, a recent innovation in artificial intelligence, has shown the ability to interpret medical images with sensitivities and specificities at or near that of skilled clinicians for some applications. In this review, we summarize the history of artificial intelligence, present some recent research advances, and speculate about the potential revolutionary clinical impact of the latest computer techniques for bone and muscle imaging. © 2019 American Society for Bone and Mineral Research. Published 2019. This article is a U.S. Government work and is in the public domain in the USA.

Artificial Intelligence Explained for Nonexperts

“…A comprehensive review of applications of deep learning in the field of musculoskeletal imaging is available in Gyftopoulos et al, 65 Hirschmann et al, 66 and Burns et al 67 For more background on the development of deep learning, LeCun and colleagues 68 provide an in-depth review. For an extensive commentary on the pitfalls and challenges of AI applications, Riley 69 and Wiens et al 70 cover different focus areas, all of which are necessary and basic knowledge for anyone who makes decisions about using AI in practice.…”

Artificial intelligence in musculoskeletal oncological radiology

“…Deep learning-based techniques are expected to minimalize false positive rates as well as assure accurate decisions and diagnoses. 49 Further automatization of radiological tasks is expected to take place in the future. Among physicians, radiologists in particular are required to perform many time-consuming tasks, like image segmentation, delineation of regions of interest, and image annotation.…”

Applications of Artificial Intelligence in Musculoskeletal Imaging: From the Request to the Report

“…That is, information other than the primary indication for the examination may be gathered from the study without need for additional radiation or exam time, thereby generating added value. 1 Conceivably, such an algorithm could evaluate all the abdominal CTs performed at an institution and provide body composition information in the report, which could be used as a clinical indicator or for epidemiological research.…”

“…It is important for radiologists to acquire expertise in AI developments in order to become leaders in their upcoming clinical implementation. Several excellent review articles have explored applications of AI in musculoskeletal (MSK) imaging, [1][2][3][4][5] including a recent issue of Seminars in Musculoskeletal Radiology dedicated to AI. 6 The present article provides an overview of the impact of AI through the entire imaging cycle of MSK radiology, from the placement of the requisition to the generation of the report, with a Canadian perspective.…”