Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

DEEP LEARNING

Detection of anaemia from retinal images

Nature Biomedical Engineering volume 4, pages 2–3 (2020)Cite this article

Subjects

A deep-learning algorithm trained with retinal images and subject metadata from the UK Biobank predicts blood-haemoglobin levels.

For narrow tasks involving the interpretation of medical images1, such as the detection of diabetic retinopathy, tuberculosis, melanoma or lymph-node metastases secondary to breast cancer2,3,4,5, deep learning has achieved diagnostic performance comparable, if not superior, to the typical performance of highly trained human specialists. Yet deep learning can also diagnose diseases that are not obvious to physicians, such as the detection of major cardiovascular risk factors (including age, gender, blood pressure and smoking status) and, surprisingly, sex (which expert ophthalmologists can predict as accurately as a coin toss), all just from retinal fundus images6,7. Now, Akinori Mitani and colleagues report in Nature Biomedical Engineering that a deep-learning algorithm trained with retinal photographs from the UK Biobank (and an enhanced version of the algorithm trained with fundus images and with ethnicity, age, sex and blood-pressure data from the same subjects) can estimate blood-haemoglobin levels (Hb levels) and thus predict the presence or absence of anaemia8.

Your institute does not have access to this article

Access options

Buy article

Get time limited or full article access on ReadCube.

$32.00

Buy

All prices are NET prices.

Fig. 1: Prediction of haemoglobin levels via deep learning.

References

  1. Ting, D. S. W. et al. Prog. Retin. Eye Res. 72, 100759 (2019).

    Article  Google Scholar 

  2. Ardila, D. et al. Nat. Med. 25, 954–961 (2019).

    CAS  Article  Google Scholar 

  3. Bejnordi, B. E. et al. JAMA 318, 2199–2210 (2017).

    Article  Google Scholar 

  4. Esteva, A. et al. Nature 542, 115–118 (2017).

    CAS  Article  Google Scholar 

  5. Ting, D. S. W. et al. JAMA 318, 2211–2223 (2017).

    Article  Google Scholar 

  6. Poplin, R. et al. Nat. Biomed. Eng. 2, 158–164 (2018).

    Article  Google Scholar 

  7. Ting, D. S. W. & Wong, T. Y. Nat. Biomed. Eng. 2, 140–141 (2018).

    Article  Google Scholar 

  8. Mitani, A. et al. Nat. Biomed. Eng. https://doi.org/10.1038/s41551-019-0487-z (2019).

    Article  Google Scholar 

  9. Yu, M. et al. Lancet Digital Health 1, e328–e329 (2019).

    Article  Google Scholar 

  10. Mannino, R. G. et al. Nat. Commun. 9, 4924 (2018).

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Singapore

    Yih-Chung Tham, Ching Yu Cheng & Tien Yin Wong

  2. Ophthalmology & Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singapore, Singapore

    Yih-Chung Tham, Ching Yu Cheng & Tien Yin Wong

Authors
  1. Yih-Chung Tham
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ching Yu Cheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Tien Yin Wong
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Tien Yin Wong.

Ethics declarations

Competing interests

Y.C.T. and C.Y.C. declare no competing interests. T.Y.W. is a co-inventor of a patent on various deep-learning systems in ophthalmology; potential conflicts of interests are managed according to institutional policies of the Singapore Health System (SingHealth) and the National University of Singapore.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Tham, YC., Cheng, C.Y. & Wong, T.Y. Detection of anaemia from retinal images. Nat Biomed Eng 4, 2–3 (2020). https://doi.org/10.1038/s41551-019-0504-2

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41551-019-0504-2

Further reading

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing