Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2018 Jun 27:6:75.
doi: 10.3389/fbioe.2018.00075. eCollection 2018.

Machine Learning in Orthopedics: A Literature Review

Federico Cabitza  1   2 , Angela Locoro  2 , Giuseppe Banfi  1
Affiliations
Review

Machine Learning in Orthopedics: A Literature Review

Federico Cabitza et al. Front Bioeng Biotechnol. .

Abstract

In this paper we present the findings of a systematic literature review covering the articles published in the last two decades in which the authors described the application of a machine learning technique and method to an orthopedic problem or purpose. By searching both in the Scopus and Medline databases, we retrieved, screened and analyzed the content of 70 journal articles, and coded these resources following an iterative method within a Grounded Theory approach. We report the survey findings by outlining the articles' content in terms of the main machine learning techniques mentioned therein, the orthopedic application domains, the source data and the quality of their predictive performance.

Keywords: deep learning; literature survey; machine learning; orthopedics; predictive models.

PubMed Disclaimer

Figures

Figure 1
Figure 1
The workflow of search and selection of papers. The syntax of the queries differs for the two search engines used, but the search terms are the same for both the searches. The queries were last executed on November 2017.
Figure 2
Figure 2
A temporal evolution chart depicting the number of papers published each year since the year 2000 and included in the survey because it discusses an ML application to Orthopedics. We distinguish between papers that discuss deep learning models and conventional ML models.
Figure 3
Figure 3
A temporal evolution chart depicting a more detailed picture than the one depicted in Figure 2. For each year, we depict the number of papers stratifying the resulting trends, using the main ML technique.
Figure 4
Figure 4
Bubble chart showing the number of papers considered in this survey arranged according to the ML techniques discussed in this section and the main classes of orthopedic application. Papers may have been counted more than once in this chart, depending on all of the ML techniques and applications reported in each of them.
Figure 5
Figure 5
Bubble chart showing the number of papers considered in this survey arranged according to data source typology and main ML class of techniques. Papers may have been counted more than once, depending on all the ML techniques and data sources exploited in each of them.
Figure 6
Figure 6
Bibliographic coupling analysis (performed by VOSviewer http://www.vosviewer.com/). The size and color of the nodes represent the number of references that are shared among the papers analyzed. The strength of a link indicates the number of cited references that two publications have in common.
Figure 7
Figure 7
Citation analysis (performed by VOSviewer http://www.vosviewer.com/). Size and brightness levels of the nodes express the number of citations of each paper (the brighter the color, the more citations of that paper). Links connect sources where one paper cites the other one (i.e., the older one).
Figure 8
Figure 8
An overview of the 70 papers analyzed in the literature review, indicating the developed ML model, the application domain and the data source typology. The papers are listed in chronological order.
Figure 9
Figure 9
Stacked bar chart representing the proportion of mentions of the evaluation metrics in the surveyed papers. A darker hue indicates a greater number of occurrences, which is also indicated in terms of percentage.
See this image and copyright information in PMC

References

    1. Abraham J. M., McCullough J. S., Parente S. T., Gaynor M. S. (2011). Prevalence of electronic health records in us hospitals. J. Healthcare Eng. 2, 121–142. 10.1260/2040-2295.2.2.121 - DOI
    1. Adankon M. M., Dansereau J., Labelle H., Cheriet F. (2012). Non invasive classification system of scoliosis curve types using least-squares support vector machines. Artif. Intell. Med. 56, 99–107. 10.1016/j.artmed.2012.07.002 - DOI - PubMed
    1. Ahmed U., Anwar A., Savage R. S., Thornalley P. J., Rabbani N. (2016). Protein oxidation, nitration and glycation biomarkers for early-stage diagnosis of osteoarthritis of the knee and typing and progression of arthritic disease. Arthr. Res. Ther. 18:250. 10.1186/s13075-016-1154-3 - DOI - PMC - PubMed
    1. Akben S. B. (2016). Importance of the shape and orientation of the spine and pelvis for the vertebral column pathologies diagnosis with using machine learning methods. Biomed. Res. S337–S342.
    1. Al-Helo S., Alomari R. S., Ghosh S., Chaudhary V., Dhillon G., Al-Zoubi M. B., et al. . (2013). Compression fracture diagnosis in lumbar: a clinical CAD system. Int. J. Comput. Assist. Radiol. Surg. 8, 461–469. 10.1007/s11548-012-0796-0 - DOI - PubMed