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
. 2021 Aug 19:9:713209.
doi: 10.3389/fcell.2021.713209. eCollection 2021.

Validation of the Relationship Between Iris Color and Uveal Melanoma Using Artificial Intelligence With Multiple Paths in a Large Chinese Population

Haihan Zhang  1 , Yueming Liu  1 , Kai Zhang  2 , Shiqi Hui  1 , Yu Feng  1 , Jingting Luo  1 , Yang Li  1 , Wenbin Wei  1
Affiliations

Validation of the Relationship Between Iris Color and Uveal Melanoma Using Artificial Intelligence With Multiple Paths in a Large Chinese Population

Haihan Zhang et al. Front Cell Dev Biol. .

Abstract

Previous studies have shown that light iris color is a predisposing factor for the development of uveal melanoma (UM) in a population of Caucasian ancestry. However, in all these studies, a remarkably low percentage of patients have brown eyes, so we applied deep learning methods to investigate the correlation between iris color and the prevalence of UM in the Chinese population. All anterior segment photos were automatically segmented with U-NET, and only the iris regions were retained. Then the iris was analyzed with machine learning methods (random forests and convolutional neural networks) to obtain the corresponding iris color spectra (classification probability). We obtained satisfactory segmentation results with high consistency with those from experts. The iris color spectrum is consistent with the raters' view, but there is no significant correlation with UM incidence.

Keywords: Chinese population; artificial intelligence; iris color; machine learning; uveal melanoma.

PubMed Disclaimer

Conflict of interest statement

KZ was employed by company SenseTime Group Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
Sex ratio of patients in the two groups.
FIGURE 2
FIGURE 2
Iris color grading of patients in two groups.
FIGURE 3
FIGURE 3
Age distribution of patients in the two groups.
FIGURE 4
FIGURE 4
The laterality of the affected eye of uveal melanoma (UM) patients.
FIGURE 5
FIGURE 5
The most representative photos in the current study population were selected as a reference for scoring and the corresponding segmentation results.
FIGURE 6
FIGURE 6
The architecture of U-Net.
FIGURE 7
FIGURE 7
Random forest. The iris color spectrum is defined as the vector in which the five probabilities ([p1,p2,p3,p4,p5]) correspond to the five grades of the color of the iris.
FIGURE 8
FIGURE 8
U-NET was used to extract iris regions from slit-lamp images, and then random forest (RF) and convolutional neural network (CNN) were used to extract iris color chromatography as descriptors to distinguish UM patients from nontumor patients. Meanwhile, the extracted iris images were directly input into the CNN network to identify UM patients and nontumor groups.
FIGURE 9
FIGURE 9
The performance for evaluating the color of iris (RF). The left-hand side is listed as the validation dataset, and the right-hand side is listed as the testing dataset. The first line shows all subjects, the second one shows the nontumor control datasets, the third one shows the tumor patient datasets.
FIGURE 10
FIGURE 10
The performance for evaluating the color of the iris (CNN).
FIGURE 11
FIGURE 11
The relationship between the color spectra of the iris and UM.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Boyle P., Day N. E., Magnus K. (1983). Mathematical modelling of malignant melanoma trends in Norway, 1953–1978. Am. J. Epidemiol. 118 887–896. 10.1093/oxfordjournals.aje.a113706 - DOI - PubMed
    1. Cerbone L., Van Ginderdeuren R., Van den Oord J., Fieuws S., Spileers W., Van Eenoo L., et al. (2014). Clinical presentation, pathological features and natural course of metastatic UM, an orphan and commonly fatal disease. Oncology 86 185–189. 10.1159/000358729 - DOI - PubMed
    1. Cumming R. G., Mitchell P., Lim R. (2000). Iris color and cataract: the blue mountains eye study. Am. J. Ophthalmol. 130 237–238. 10.1016/s0002-9394(00)00479-7 - DOI - PubMed
    1. Damato E. M., Damato B. E. (2012). Detection and time to treatment of UM in the United Kingdom: an evaluation of 2,384 patients. Ophthalmology 119 1582–1589. 10.1016/j.ophtha.2012.01.048 - DOI - PubMed
    1. Dogrusöz M., Jager M. J., Damato B. (2017). UM treatment and prognostication. Asia Pac. J. Ophthalmol. 6 186–196. 10.22608/APO.201734 - DOI - PubMed

LinkOut - more resources