The clinicians' perspectives on machine learning - Nature.com

jokbanga.blogspot.com

The intended purpose of machine learning (ML) in cardiovascular medicine is to help guide clinical diagnoses as well as promote scientific discovery. Whether ML is implemented by most clinical cardiologists and cardiovascular researchers will likely depend on the successful resolution of concerns fueling hesitancy to embrace ML. This commentary discusses caveats related to ML in clinical practice, and offers suggestions for stakeholders on how to bridge knowledge gaps and clinicians’ misgivings to bring this powerful approach to the clinic to improve care of the patients we serve.

First is a transparency issue. Most clinicians view ML as a black box. Undoubtedly, this leads them to question whether ML can be used to make a decision that may impact life or death. This skepticism is further intensified as most ML developers indeed admit that ML models, such as a deep learning model, can be hard to interpret despite various algorithms developed to tackle this problem. In addition, sometimes ML unveils an entirely new link between biological features and patient outcomes. A great example is a recent publication using deep learning to demonstrate that branching complexity and density of retinal microvasculature is associated with higher risk of cardiovascular disease (CVD)². As this novel finding has not been validated by any pre-clinical model, the question remains, can we trust the ML-based discoveries not yet corroborated in practice?

This is a preview of subscription content

Access options

Access through your institution

Change institution

Buy or subscribe

Subscribe to Journal

Get full journal access for 1 year

We are sorry, but there is no personal subscription option available for your country.

More Information

Buy article

Get time limited or full article access on ReadCube.

$32.00

Buy

All prices are NET prices.

Additional access options:

• Log in
• Learn about institutional subscriptions

Fig. 1: Characteristics and applications of machine learning in cardiovascular medicine.

References

1. Benjamens, S., Dhunnoo, P. & Meskó, B. npj Digit. Med. 3, 118 (2020).

   Article  Google Scholar 

2. Zekavat, S. M. et al. Circulation 145, 134–150 (2022).

   Article  Google Scholar 

3. Zech, J. R. et al. PLoS Med. 15, e1002683 (2018).

   Article  Google Scholar 

4. Reardon, S. Nature 576, S54–S58 (2019).

   CAS  Article  Google Scholar 

5. Huisman, M. et al. Eur. Radiol. 31, 7058–7066 (2021).

   Article  Google Scholar 

6. Nikpay, M. et al. Nat. Genet. 47, 1121–1130 (2015).

   CAS  Article  Google Scholar 

7. Pattarabanjird, T. et al. Genes 11, 1446 (2020).

   CAS  Article  Google Scholar 

8. Ridker, P. M. et al. N. Engl. J. Med. 377, 1119–1131 (2017).

   CAS  Article  Google Scholar 

Download references

Author information

Affiliations

1. Carter Immunology Center, University of Virginia, Charlottesville, VA, USA

   Tanyaporn Pattarabanjird & Coleen McNamara

2. Cardiovascular Research Center, University of Virginia, Charlottesville, VA, USA

   Tanyaporn Pattarabanjird & Coleen McNamara

3. Department of Biomedical Engineering University of Virginia, Charlottesville, VA, USA

   Tanyaporn Pattarabanjird

4. Division of Cardiovascular Medicine/Department of Medicine, University of Virginia, Charlottesville, VA, USA

   Coleen McNamara

Authors

1. Tanyaporn Pattarabanjird
   View author publications

   You can also search for this author in PubMed Google Scholar

2. Coleen McNamara
   View author publications

   You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Coleen McNamara.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Pattarabanjird, T., McNamara, C. The clinicians’ perspectives on machine learning. Nat Cardiovasc Res (2022). https://ift.tt/nqbiu7G

Download citation

• Published: 28 February 2022

• DOI: https://ift.tt/nqbiu7G

Adblock test (Why?)

"machine" - Google News
February 28, 2022 at 11:04PM
https://ift.tt/8OIVYjW

The clinicians' perspectives on machine learning - Nature.com
"machine" - Google News
https://ift.tt/2EC6Ncj
https://ift.tt/B7LR2Az

Bagikan Berita Ini