This site needs JavaScript to work properly. Please enable it to take advantage of the complete set of features!
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.

NIH NLM Logo
Log in
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Editorial
. 2021 Feb 11;17(2):e1008531.
doi: 10.1371/journal.pcbi.1008531. eCollection 2021 Feb.

Ten simple rules for engaging with artificial intelligence in biomedicine

Affiliations
Editorial

Ten simple rules for engaging with artificial intelligence in biomedicine

Avni Malik et al. PLoS Comput Biol. .
No abstract available

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Understand what Artificial Intelligence is.
Data analysis techniques have become more advanced over time. The elaboration of artificial intelligence techniques influenced machine learning. Similarly, the sophistication of machine learning prompted evolution into deep learning as shown in this figure adapted from Le Berre and colleagues [13]. Each approach shows an example of its application in the biomedical field.
Fig 2
Fig 2. Understand what Artificial Intelligence is.
The predictive behaviors of AI models can sometimes be hidden from programmers, who must then view the internal workings as a "black box," analyzing them based on their inputs and outputs.
Fig 3
Fig 3. Try to peek into the "black box" where you can.
The recognition of objects in images done by Grad-CAMs relies on classification of these objects by programmers. Without the explicit knowledge of what "clocks," "toilets and sinks," giraffes, or "stop signs" are, the programmer would not be able to verify the outputs of the Grad-CAM visualization. Mastery of the topic of interest is crucial in the development of such models [19].
Fig 4
Fig 4. Try to peek into the "black box" where you can.
With the help of experts in gastrointestinal tissues, Grad-CAMs can be trained to recognize many different key features (specified by the blue color) in biomedical images to aid with celiac disease classification [20].
Fig 5
Fig 5. Find the potential of AI in practice.
The Remidio "Fundus on Phone" has reformed fundoscopy by creating more accessible and economical technology to optimize healthcare [22].

References

    1. Topol E. Deep medicine: how artificial intelligence can make healthcare human again. New York, New York: Basic Books; 2019.
    1. Smith TM. 10 ways health care AI could transform primary care. American Medical Association; 2020. [cited 2020 May 17]. Available from: https://www.ama-assn.org/practice-management/digital/10-ways-health-care....
    1. Zemouri R, Zerhouni N, Racoceanu D. Deep Learning in the biomedical applications: recent and future status. Appl Sci. 2019. April;9 10.3390/app9081526 - DOI
    1. Nicholson W. Risks and remedies for artificial intelligence in health care. Brookings. Brookings; 2020. [cited 2020 May 17]. Available from: https://www.brookings.edu/research/risks-and-remedies-for-artificial-int....
    1. News Center. Stanford Medicine launches health care trends report. News Center. 2017. [cited 2020 May 17]. Available from: http://med.stanford.edu/news/all-news/2017/06/stanford-medicine-launches....

Publication types

LinkOut - more resources

Full text links
Public Library of Science full text link Public Library of Science Free PMC article
Cite

AltStyle によって変換されたページ (->オリジナル) /