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
. 2024 Aug;102(5):e831-e841.
doi: 10.1111/aos.16616. Epub 2023 Dec 22.

Suitability of machine learning for atrophy and fibrosis development in neovascular age-related macular degeneration

Jesus de la Fuente  1   2 , Sara Llorente-González  3   4   5 , Patricia Fernandez-Robredo  3   4   5 , María Hernandez  3   4   5 , Alfredo García-Layana  3   4   5 , Idoia Ochoa  1   6 , Sergio Recalde  3   4   5 ; Spanish AMD group
Collaborators, Affiliations
Free article

Suitability of machine learning for atrophy and fibrosis development in neovascular age-related macular degeneration

Jesus de la Fuente et al. Acta Ophthalmol. 2024 Aug.
Free article

Abstract

Purpose: To assess the suitability of machine learning (ML) techniques in predicting the development of fibrosis and atrophy in patients with neovascular age-related macular degeneration (nAMD), receiving anti-VEGF treatment over a 36-month period.

Methods: An extensive analysis was conducted on the use of ML to predict fibrosis and atrophy development on nAMD patients at 36 months from start of anti-VEGF treatment, using only data from the first 12 months. We use data collected according to real-world practice, which includes clinical and genetic factors.

Results: The ML analysis consistently identified ETDRS as a relevant factor for predicting the development of atrophy and fibrosis, confirming previous statistical analyses. Also, it was shown that genetic variables did not demonstrate statistical relevance in the prediction. Despite the complexity of predicting macular degeneration, our model was able to obtain a balance accuracy of 63% and an AUC of 0.72 when predicting the development of atrophy or fibrosis at 36 months.

Conclusion: This study demonstrates the potential of ML techniques in predicting the development of fibrosis and atrophy in nAMD patients receiving long-term anti-VEGF treatment. The findings highlight the importance of clinical factors, particularly ETDRS (early treatment diabetic retinopathy study) visual acuity test, in predicting these outcomes. The lessons learned from this research can guide future ML-based prediction tasks in the field of ophthalmology and contribute to the design of data collection processes.

Keywords: atrophy; extreme gradient boosting; fibrosis; machine learning; nAMD; random Forest; support vector machine.

PubMed Disclaimer

References

REFERENCES

    1. Ashraf, M., Souka, A. & Adelman, R.A. (2018) Age‐related macular degeneration: using morphological predictors to modify current treatment protocols. Acta Ophthalmologica, 96(2), 120–133.
    1. Bhisitkul, R.B., Mendes, T.S., Rofagha, S., Enanoria, W., Boyer, D.S., Sadda, S.V.R. et al. (2015) Macular atrophy progression and 7‐year vision outcomes in subjects from the ANCHOR, MARINA, and HORIZON studies: the SEVEN‐UP study. American Journal of Ophthalmology, 159(5), 915–924.
    1. Brantley, M.A., Jr., Fang, A.M., King, J.M., Tewari, A., Kymes, S.M. & Shiels, A. (2007) Association of complement factor H and LOC387715 genotypes with response of exudative age‐related macular degeneration to intravitreal bevacizumab. Ophthalmology, 114(12), 2168–2173.
    1. Breiman, L. (2001) Random forests. Machine Learning, 45(1), 5–32.
    1. Caire, J., Recalde, S., Velazquez‐Villoria, A., Garcia‐Garcia, L., Reiter, N., Anter, J. et al. (2014) Growth of geographic atrophy on fundus autofluorescence and polymorphisms of CFH, CFB, C3, FHR1‐3, and ARMS2 in age‐related macular degeneration. JAMA Ophthalmology, 132(5), 528–534.

MeSH terms

Substances

LinkOut - more resources