Most humans have trichromatic vision: they perceive blue, green, and red colors using three types of retinal photoreceptor cells that are sensitive to short, medium, or long wavelengths of light, respectively. Color-vision deficiency (CVD) affects between 2% and 8% of males (depending on ancestry) and approximately 0.5% of females (Delpero et al., 2005). Congenital CVD is commonly caused by mutations in the genes (or nearby promoter regions) that code for red or green cone photopigments; these genes are proximal to each other on the X chromosome (Nathans et al., 1986).

CVD is divided into categories, the most common being deutan CVD, affecting approximately 6% of males of European descent, and protan CVD, affecting 2% of males of European descent (Delpero et al., 2005). Both categories are commonly known as red-green colorblindness. Within each category, CVD is subclassified according to whether individuals are dichromats—able to see two primary colors—or anomalous trichromats—able to see three primary colors but differently from normal trichromats. Anomalous trichromats differ in the degree of severity with which they can distinguish color patterns. Individuals with deuteranopia (‘deuteranopes’) or protanopia do not have corresponding green or red cones, respectively (Simunovic, 2010). Individuals with deuteranomaly do not have properly functioning green cones, and those with protanomaly do not have properly functioning red cones. People with any of these conditions often see green and red as brown or beige colors. Thus, when images contain shades of green and red—or when either is paired with brown—parts of the image may be indistinguishable. Furthermore, it can be problematic when some pinks or oranges are paired with greens. These issues can lead to incorrect interpretations of figures in scientific journal articles for individuals with CVD.

Efforts have been made to ensure that scientific figures are accessible to people with CVD. For example, researchers have developed algorithms that attempt to recolor images so that people with CVD can more easily interpret them (Flatla, 2011; Lin et al., 2019; Tsekouras et al., 2021). However, these tools are not in wide use, and more work is needed to verify their efficacy in practice. In the meantime, as researchers prepare scientific figures, they can take measures to improve accessibility for people with CVD. For example, they can avoid rainbow color maps that show colors in a gradient; they can use color schemes or color intensities that are CVD friendly (Crameri et al., 2020); additionally, they can provide labels that complement information implied by color differences. However, for the millions of images that have already been published in scientific articles, little is known about the frequency with which these images are CVD friendly. The presence or absence of particular color pairings—and distances between them—can be quantified computationally to estimate this frequency. However, a subjective evaluation of individual images is necessary to identify whether color pairings and distances are likely to affect scientific interpretation.

In this paper, we focus on deuteranopia and its subtypes. To estimate the extent to which the biological and medical literature contains images that may be problematic to deuteranopes, we manually reviewed a ‘training set’ of 4964 images and two ‘test sets’ of 879 and 1191 images, respectively. These images were published in articles between the years 2012 and 2022. After identifying images that we deemed most likely to be problematic or not, we used machine-learning algorithms to identify patterns that could discriminate between these two categories of images and thus might be useful for automating the identification of problematic images. If successful, such an algorithm could be used to alert authors, presenters, and publishers that scientific images could be modified to improve visual accessibility and thus make biological and medical fields more inclusive.

We downloaded images from research articles published in the eLife journal. Not counting duplicate versions of the same image, we obtained 66,253 images. Of these images, 1744 (2.6%) were grayscale (no color). Of these images, 56,816 (85.6%) included at least one color pair for which the amount of contrast might be problematic to people with moderate-to-severe deuteranopia (‘deuteranopes’). To characterize potentially problematic aspects of each color-based image, we calculated five metrics based on color contrasts and distances; we also compared the color profiles against what deuteranopes might see. The mean, pixel-wise color distance between the original and simulated image exhibited a bimodal distribution, according to Hartigans’ Dip Test for Unimodality (p<0.001; Hartigan and Hartigan, 1985). Specifically, 4708 images (7.3%) had a difference smaller than 0.01, while the median difference for the remaining images was 0.05 (Figure 1). Most other metrics showed similar patterns, although bimodality was less apparent through visual inspection (Figure 2; Figure 3; Figure 4). The exception was the proportion of pixels in the original image that used a color from one of the high-ratio color pairs, which was unimodal (p=1; Figure 5).

Figure 1

Download asset Open asset

Figure 2

Download asset Open asset

Figure 3

Download asset Open asset

Figure 4

Download asset Open asset

Figure 5

Download asset Open asset

We determined that many images with the highest (or lowest, as would be the case for the ‘Mean Euclidean distance between pixels for high-ration color pairs’) scores for these metrics would be problematic for deuteranopes. However, we noted that certain color pairs were more problematic than others, and the use of effective labels and/or spacing between colors often mitigated potential problems. Thus, to better estimate the extent to which images are problematic for deuteranopes, we manually reviewed a sample of 4964 images and judged whether it would be likely for deuteranopes to recognize the scientific message behind each image. Supplementary file 2 contains a record of these evaluations, along with comments that indicate either problematic aspects of the images or factors that mitigated potential problems. We concluded that 636 (12.8%) of the images were ‘Definitely problematic’, whereas 3865 of the images (77.9%) were ‘Definitely okay’. The remaining images were grayscale (n=179), or we were unable to reach a confident conclusion (n=284). For the images that were ‘Definitely okay’, we visually detected shades of green and red or orange in 2348 (60.8%) images; however, in nearly all (99.3%) of these cases, we deemed that the contrasts between the shades were sufficient that a deuteranope could interpret the images. Furthermore, distance between the colors and/or labels within the images mitigated potential problems in 54.2% and 48.4% of cases, respectively.

We evaluated longitudinal trends and differences among biology subdisciplines. In some cases for the eLife articles, multiple images came from the same journal article. Therefore, to avoid pseudoreplication, we categorized each article as either ‘Definitely okay’ or ‘Definitely problematic’. If an article included at least one ‘Definitely problematic’ image, we categorized the entire article under this category. The percentage of ‘Definitely problematic’ articles declined steadily between 2012 and 2021, with a modest increase in 2022 (Figure 6). (Fewer articles were available for 2022 than for prior years.) Using a generalized linear model with a binomial family to perform logistic regression, we found this decline to be statistically significant (p<0.001). A χ² goodness-of-fit test revealed that the number of ‘Definitely problematic’ articles differed significantly by subdiscipline (p<0.001). The subdisciplines with the highest percentages of problematic articles were Cell Biology, Developmental Biology, and Stem Cells and Regenerative Medicine (Figure 7). The subdisciplines with the lowest percentages of problematic articles were Evolutionary Biology, Genetics and Genomics, and Computational and Systems Biology.

Figure 6

Download asset Open asset

Figure 7

Download asset Open asset

Despite the benefits of manual review, this process is infeasible on a large scale. Therefore, we evaluated techniques for automating image classification. As an initial test, we used the five image-quantification metrics. We also combined these into a single, ranked-based score for each image. In all cases, the metrics differed significantly between the ‘Definitely okay’ and ‘Definitely problematic’ images (Figure 8; Figure 9; Figure 10; Figure 11; Figure 12; Figure 13). To estimate their predictive performance, we performed cross validation on the training set. Values relatively close to 1.0 indicate relatively high performance. A value of 0.5 indicates that predictions are no better than random guessing. The best-performing metric was number of color pairs that exhibited a high color-distance ratio between the original and simulated images (AUROC: 0.75; AUPRC: 0.34). All the other metrics—except mean, pixel-wise color distance between the original and simulated image—performed better than random guessing (Supplementary file 1A). As an alternative to the combined rank score, we used classification algorithms to make predictions with the five metrics as inputs. In cross-validation on the training set, the best-performing algorithm was Logistic Regression (AUROC: 0.82; AUPRC: 0.43; Supplementary file 1B).

Figure 8

Download asset Open asset

Figure 9

Download asset Open asset

Figure 10

Download asset Open asset

Figure 11

Download asset Open asset

Figure 12

Download asset Open asset

Figure 13

Download asset Open asset

Additionally, we created a convolutional neural network (CNN) to make predictions according to visual and spatial patterns in the images. CNNs are highly configurable and often sensitive to model parameters and configurations. Accordingly, we performed multiple iterations of cross validation on the training set and compared a variety of hyperparameters and configurations. All non-default options performed better than Logistic Regression based on the image-quantification metrics (Supplementary file 1C). The best-performing approach used class weighting; early stopping; random flipping and rotation (threshold: 0.2); a dropout rate of 0.5; the ResNet pre-trained model for transfer learning; and model fine tuning.

We manually reviewed a hold-out test set that consisted of 1,000 additional images from eLife (Supplementary file 3). After we removed images that were not ‘Definitely okay’ or ‘Definitely problematic’, 879 images remained. For the Logistic Regression algorithm and CNN, we trained models using the full training set and classified each hold-out image as ‘Definitely okay’ or ‘Definitely problematic’. Logistic Regression classified the images with an AUROC of 0.82 (AUPRC: 0.49; Figure 14; Figure 15; Figure 16). The CNN classified the images with an AUROC of 0.89 (AUPRC: 0.77; Figure 17; Figure 18; Figure 19). Supplementary file 4 indicates the performance of both models for a variety of classification metrics.

Figure 14

Download asset Open asset

Figure 15

Download asset Open asset

Figure 16

Download asset Open asset

Figure 17

Download asset Open asset

Figure 18

Download asset Open asset

Figure 19

Download asset Open asset

For the 92 hold-out images that were misclassified by the CNN model, we compared them against our manual annotations and determined that in 13 cases, the reviewers had missed subtle patterns; we conclude that it would be justified to change these labels (Supplementary file 5). For 31 of the misclassified images, we visually identified patterns that might have confused the CNN; however, upon reevaluation, we maintain that the original labels were valid. For the remaining 48 misclassified images, we were unable to identify patterns that seemed likely to have confused the model.

Lastly, we used the Logistic Regression and CNN models to predict ‘Definitely okay’ or ‘Definitely problematic’ status for a second hold-out set with images from PubMed Central (Supplementary file 6). Of the non-grayscale images, 104 were labeled as ‘Definitely problematic’ and 1087 as ‘Definitely okay’. Again, the CNN model (AUROC: 0.78; AUPRC: 0.39; Figure 20; Figure 21; Figure 22) outperformed the Logistic Regression model (AUROC: 0.73; AUPRC: 0.16; Figure 23; Figure 24; Figure 25; Supplementary file 4).

Figure 20

Download asset Open asset

Figure 21

Download asset Open asset

Figure 22

Download asset Open asset

Figure 23

Download asset Open asset

Figure 24

Download asset Open asset

Figure 25

Download asset Open asset

To learn more about the prevalence of images in the biology literature that may be inaccessible to deuteranopes, we examined thousands of articles from the eLife journal. eLife uses a content-licensing scheme that made it possible to perform this study in a transparent manner. Additionally, we selected articles from PubMed Central, as images from these articles represent life-science journals more broadly. After manual review, we estimate that 12.8% of the figures in eLife would be challenging to interpret for scientists with moderate-to-severe deuteranopia. The percentage of ‘Definitely problematic’ figures in PubMed Central articles was considerably lower (5.2%). One reason is that a much higher percentage (38.5%) of the images from PubMed Central were grayscale compared to those from eLife (4.4%). The findings for both sources indicate that color accessibility is a problem for thousands of journal articles per year.

Significant work has been done to address and improve accessibility for individuals with CVD (Zhu and Mao, 2021). This work can be categorized into four types of studies: simulation methods, recolorization methods, estimating the frequency of accessible images, and educational. Simulation methods have been developed to better understand how images appear to individuals with CVD. Brettel et al. first simulated CVDs using the long, medium, and short (LMS) colorspace (Brettel et al., 1997). For dichromacy, the colors in the LMS space are projected onto an axis that corresponds to the non-functional cone cell. Viénot et al. expanded on this work by applying a 3x3 transformation matrix to simulate images in the same LMS space (Vinot et al., 1999). Machado et al. created matrices to simulate CVDs based on the shift theory of cone cell sensitivity (Machado et al., 2009; Stockman and Sharpe, 2000). These algorithms allow individuals without CVD to qualitatively test how their images might appear to people with CVD. The simulation algorithms and matrices are freely available and accessible via websites and software packages (Coblis, 2021; Color blind, 2020; DaltonLens, 2023; Wilke, 2023).

CVD simulations have facilitated the creation of colorblind-friendly palettes (Olson and Brewer, 1997), and they have led to algorithms that recolor images to become more accessible to people with CVD. Recolorization methods focus on enhancing color contrasts and preserving image naturalness (Zhu and Mao, 2021). Many algorithms have been developed to compensate for dichromacy (Jefferson and Harvey, 2007; Huang et al., 2007; Ruminski et al., 2010; Rasche et al., 2005; Machado and Oliveira, 2010; Ching and Sabudin, 2010; Ribeiro and Gomes, 2020; Li et al., 2020; Wang et al., 2021; Nakauchi and Onouchi, 2008; Zhu et al., 2019b; Ma et al., 2009). These algorithms apply a variety of techniques including hue rotation, customized difference addition, node mapping, and generative adversarial networks (Zhu and Mao, 2021; Li et al., 2020). Many of these methods have been tested for efficacy, both qualitatively and quantitatively (Zhu and Mao, 2021). Recolorization algorithms have been applied to PC displays, websites, and smart glasses (Tanuwidjaja, 2014). Despite the prevalence of these algorithms, current techniques have not been systematically compared and may sacrifice image naturalness to increase contrast. Additionally, recoloring may not improve the accessibility of some scientific figures because papers often reference colors in figure descriptions; recoloring the image could interfere with matching colors between the text and images.

An increase in available resources for making figures accessible to individuals with CVD has prompted some researchers to investigate whether these resources have been impactful in decreasing the frequency of publishing scientific figures with problematic colors. Frane examined the prevalence of images in psychology journals that could be confusing to people with CVD (Frane, 2015). A group of panelists with CVD qualitatively evaluated 246 images and found that 13.8% of color figures caused difficulty for at least one panelist; this percentage is similar to our findings. They also found that in instructions to authors, journals rarely mentioned the importance of designing figures for CVD accessibility. Angerbauer et al. recruited crowdworkers to analyze a sample of 1,710 published images and to identify issues with the use of color (Angerbauer et al., 2022). On average, 60% of the sampled images were given a rating of ‘accessible’ across CVD types. From 2000 to 2019, they observed a slight increase in CVD accessibility for published figures.

Educational resources are available to researchers looking to make their figures suitable for people with CVD. For example, Jambor et al. provide guidelines and examples to help researchers avoid common problems (Jambor et al., 2021). JetFighter scans preprints from bioRxiv and searches for rainbow-based color schemes (Saladi and Maggiolo, 2019). When these are identified, JetFighter notifies the authors about page(s) that might need to be adjusted. However, as we have shown, the presence of particular color combinations does not necessarily indicate that an image is problematic to people with deuteranopia. Frequently, a more nuanced evaluation is necessary.

The seaborn Python package includes a ‘colorblind’ palette (Waskom, 2021). The colorBlindness package for R provides simulation tools and CVD-friendly palettes (Ou, 2021). The scatterHatch package facilitates creation of CVD-friendly scatter plots for single-cell data (Guha et al., 2022). When designing figures, researchers may find it useful to first design them so that key elements are distinguishable in grayscale. Then, color can be added—if necessary—to enhance the image. Color should not be used for the sole purpose of making an image aesthetically pleasing. Using minimal color avoids problems that arise from color pairing issues. Rainbow color maps, in particular, should be avoided. If a researcher finds it necessary to include problematic color pairings in figures, they can vary the saturation and intensity of the colors so they are more distinguishable to people with CVD. Many of the problematic figures that we identified in this study originated from fluorescence microscopy experiments, where red and green dyes were used. Choosing alternative color dyes could reduce this problem and improve the interpretability of microscopy images for people in all fields.

Our analysis has limitations. Firstly, it relied on deuteranopia simulations rather than the experiences of deuteranopes. However, by using simulations, the reviewers were capable of seeing two versions of each image: the original and a simulated version. We believe this is important in assessing the extent to which deuteranopia could confound image interpretations. Conceivably, this could be done with deuteranopes after recoloration, but it is difficult to know whether deuteranopes would see the recolored images in the same way that non-deuteranopes see the original images. Secondly, because we used a single, relatively high severity threshold, our simulations do not represent the full spectrum of experiences that scientists with deuteranopia have. Thus, our findings and tools should be relevant to some (but not all) people with deuteranopia. Furthermore, recent evidence suggests that commonly used mathematical representations of color differences are unlikely to reflect human perceptions perfectly (Bujack et al., 2022). As methods evolve for more accurately simulating color perception, we will be more capable of estimating the extent to which scientific figures are problematic for deuteranopes. Thirdly, our evaluations focused on deuteranopia, the most common form of CVD. It will be important to address other forms of CVD, such as protanopia, in future work. Fourthly, we identified some images as ‘Probably problematic’ or ‘Probably okay’. Using our review process, we were unable to draw firm conclusions about these images. To avoid adding noise to the classification analyses—we excluded these images and provided notes reflecting our reasoning. Future work may help to clarify these labels. Finally, our CNN model performed well at differentiating between ‘Definitely okay’ and ‘Definitely problematic’ images in the eLife hold-out test set; however, the model’s predictive performance dropped considerably when applied to the PubMed Central hold-out test set. Many of the eLife images are from cell-related research, and we labeled many of these as problematic. Many other image types were also identified as unfriendly, including heat maps, line charts, maps, three-dimensional structural representations of proteins, photographs, network diagrams, etc. Our model may have developed a bias toward patterns specific to image types that are over-represented in eLife, affecting its performance for other journals. The PubMed Central Open Access Subset contains articles for thousands of journals, spanning diverse subdisciplines of biology and medicine. It seems likely that this diversity is a factor behind the drop in performance. Future efforts to review larger collections of PubMed Central articles could help to overcome this limitation.

By summarizing color patterns in more than 66,000 images and manually reviewing 8000 images, we have created an open data resource that other researchers can use to develop their own methods. Using all of these images, we trained a machine-learning model that predicts whether images are friendly to deuteranopes. It is available as a Web application (https://bioapps.byu.edu/colorblind_image_tester). Scientists and others can use it to obtain insights into whether individual images are accessible to scientists with deuteranopia. However, this tool should be used as a starting point only. Human judgment remains essential.

The images we used for evaluation and the trained TensorFlow models are stored in an Open Science Framework repository (https://osf.io/8yrkb). It also includes folders with images marked as "friendly" or "unfriendly" to facilitate examination of images in either category. The code for processing and analyzing the images is available at GitHub (copy archived at Piccolo and Stevens, 2024). That repository also includes the calculated metrics, cross-validation assignments, results of image curation, and outputs of the classification algorithms. The Web application code is available at GitHub (copy archived at Piccolo, 2024). The code for processing and analyzing the images is available as a Zenodo archive (https://doi.org/10.5281/zenodo.13366997). It includes the calculated metrics, cross-validation assignments, results of image curation, and outputs of the classification algorithms. The Web application code is available as a Zenodo archive (https://doi.org/10.5281/zenodo.13367011).

1. 1. Piccolo SR

   (2023) Open Science Framework

   Supplementary images from eLife articles and the PubMed Open Access Data Subset.

   https://doi.org/10.17605/OSF.IO/8YRKB

1. Conference

   1. Abadi M

   (2016)

   TensorFlow{}: A system for {large-scale} machine learning

   in 12th USENIX symposium on operating systems design and implementation (OSDI. pp. 265–283.

   • Google Scholar
2. Book

   1. Aisch G

   (2018)

   I wrote some code that automatically checks visualizations for non-colorblind safe colors

   Here’s How It Works.

   • Google Scholar
3. Conference

   1. Angerbauer K
   2. Rodrigues N
   3. Cutura R
   4. Öney S
   5. Pathmanathan N
   6. Morariu C
   7. Weiskopf D
   8. Sedlmair M

   (2022) Accessibility for color vision deficiencies: challenges and findings of a large scale study on paper figures

   CHI ’22.

   https://doi.org/10.1145/3491102.3502133

   • Google Scholar
4. 1. Breiman L

   (2001) Random forests

   Machine Learning 45:5–32.

   https://doi.org/10.1023/A:1010933404324

   • Google Scholar
5. 1. Brettel H
   2. Viénot F
   3. Mollon JD

   (1997) Computerized simulation of color appearance for dichromats

   Journal of the Optical Society of America A 14:2647.

   https://doi.org/10.1364/JOSAA.14.002647

   • Google Scholar
6. 1. Bujack R
   2. Teti E
   3. Miller J
   4. Caffrey E
   5. Turton TL

   (2022) The non-Riemannian nature of perceptual color space

   PNAS 119:e2119753119.

   https://doi.org/10.1073/pnas.2119753119

   • PubMed
   • Google Scholar
7. 1. Chen W
   2. Chen W
   3. Bao H

   (2011) An efficient direct volume rendering approach for dichromats

   IEEE Transactions on Visualization and Computer Graphics 17:2144–2152.

   https://doi.org/10.1109/TVCG.2011.164

   • PubMed
   • Google Scholar
8. Conference

   1. Ching SL
   2. Sabudin M

   (2010) 2010 2nd International Conference on Computer Technology and Development (ICCTD)

   Cairo, Egypt.

   https://doi.org/10.1109/ICCTD.2010.5645874

   • Google Scholar
9. Website

   1. Coblis

   (2021) Coblis — Color Blindness Simulator – Colblindor

   Accessed June 1, 2023.

   https://www.color-blindness.com/coblis-color-blindness-simulator/#google_vignette
10. Book

    1. Color blind

    (2020)

    Color Blind Safe Colors on Color Wheel Adobe Color

    Adobe Color.

    • Google Scholar
11. Software

    1. Corporation M
    2. Weston S

    (2022) DoParallel: foreach parallel adaptor for the ’Parallel’ package

    R-Forge.

    https://rdrr.io/rforge/doParallel/
12. 1. Crameri F
    2. Shephard GE
    3. Heron PJ

    (2020) The misuse of colour in science communication

    Nature Communications 11:5444.

    https://doi.org/10.1038/s41467-020-19160-7

    • PubMed
    • Google Scholar
13. Software

    1. DaltonLens

    (2023) DaltonLens-python, version 3cba5e6

    Github.

    https://github.com/DaltonLens/DaltonLens-Python
14. Software

    1. Davis G

    (2022) SpacesXYZ: CIE XYZ and some of its derived color spaces, version 1.3-0

    CRAN.

    https://CRAN.R-project.org/package=spacesXYZ
15. 1. Delpero WT
    2. O’Neill H
    3. Casson E
    4. Hovis J

    (2005)

    Aviation-relevent epidemiology of color vision deficiency

    Aviation, Space, and Environmental Medicine 76:127–133.

    • PubMed
    • Google Scholar
16. 1. Fix E
    2. Hodges JL

    (1989) Discriminatory analysis. nonparametric discrimination: consistency properties

    International Statistical Review / Revue Internationale de Statistique 57:238–247.

    https://doi.org/10.2307/1403797

    • Google Scholar
17. Conference

    1. Flatla DR

    (2011) Accessibility for individuals with color vision deficiency

    UIST ’11, Santa Barbara California USA.

    https://doi.org/10.1145/2046396.2046412

    • Google Scholar
18. 1. Frane A

    (2015) A call for considering color vision deficiency when creating graphics for psychology reports

    The Journal of General Psychology 142:194–211.

    https://doi.org/10.1080/00221309.2015.1063475

    • PubMed
    • Google Scholar
19. Book

    1. Géron A

    (2022)

    Hands-on machine learning with scikit-learn, keras, and tensorflow

    O’Reilly Media, Inc.

    • Google Scholar
20. 1. Grau J
    2. Grosse I
    3. Keilwagen J

    (2015) PRROC: computing and visualizing precision-recall and receiver operating characteristic curves in R

    Bioinformatics 31:2595–2597.

    https://doi.org/10.1093/bioinformatics/btv153

    • PubMed
    • Google Scholar
21. 1. Guha T
    2. Fertig EJ
    3. Deshpande A

    (2022) Generating colorblind-friendly scatter plots for single-cell data

    eLife 11:e82128.

    https://doi.org/10.7554/eLife.82128

    • PubMed
    • Google Scholar
22. 1. Hartigan JA
    2. Hartigan PM

    (1985) The dip test of unimodality

    The Annals of Statistics 13:1176346577.

    https://doi.org/10.1214/aos/1176346577

    • Google Scholar
23. Conference

    1. He K
    2. Zhang X
    3. Ren S
    4. Sun J

    (2016) Deep Residual Learning for Image Recognition

    2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 770–778.

    https://doi.org/10.1109/CVPR.2016.90

    • Google Scholar
24. 1. Huang JB
    2. Tseng YC
    3. Wu SI
    4. Wang SJ

    (2007) Information preserving color transformation for protanopia and deuteranopia

    IEEE Signal Processing Letters 14:711–714.

    https://doi.org/10.1109/LSP.2007.898333

    • Google Scholar
25. Book

    1. Hunt RWG

    (2005)

    The Reproduction of Colour

    John Wiley & Sons.

    • Google Scholar
26. 1. Jambor H
    2. Antonietti A
    3. Alicea B
    4. Audisio TL
    5. Auer S
    6. Bhardwaj V
    7. Burgess SJ
    8. Ferling I
    9. Gazda MA
    10. Hoeppner LH
    11. Ilangovan V
    12. Lo H
    13. Olson M
    14. Mohamed SY
    15. Sarabipour S
    16. Varma A
    17. Walavalkar K
    18. Wissink EM
    19. Weissgerber TL

    (2021) Creating clear and informative image-based figures for scientific publications

    PLOS Biology 19:e3001161.

    https://doi.org/10.1371/journal.pbio.3001161

    • PubMed
    • Google Scholar
27. Conference

    1. Jefferson L
    2. Harvey R

    (2007) An interface to support color blind computer users

    CHI07, San Jose California USA.

    https://doi.org/10.1145/1240624.1240855

    • Google Scholar
28. Conference

    1. Krizhevsky A
    2. Sutskever I
    3. Hinton GE

    (2012)

    Imagenet classification with deep convolutional neural networks

    Advances in neural information processing systems 25.

    • Google Scholar
29. 1. Kuhn GR
    2. Oliveira MM
    3. Fernandes LAF

    (2008) An improved contrast enhancing approach for color-to-grayscale mappings

    The Visual Computer 24:505–514.

    https://doi.org/10.1007/s00371-008-0231-2

    • Google Scholar
30. Software

    1. Kuhn M
    2. Vaughan D
    3. Hvitfeldt E

    (2023) Yardstick: tidy characterizations of model performance, version 1.3.1

    Tidy Data.

    https://yardstick.tidymodels.org/
31. 1. Li H
    2. Zhang L
    3. Zhang X
    4. Zhang M
    5. Zhu G
    6. Shen P
    7. Li P
    8. Bennamoun M
    9. Shah SAA

    (2020) Color vision deficiency datasets & recoloring evaluation using GANs

    Multimedia Tools and Applications 79:27583–27614.

    https://doi.org/10.1007/s11042-020-09299-2

    • Google Scholar
32. 1. Lin HY
    2. Chen LQ
    3. Wang ML

    (2019) Improving discrimination in color vision deficiency by image re-coloring

    Sensors 19:2250.

    https://doi.org/10.3390/s19102250

    • PubMed
    • Google Scholar
33. 1. Ma Y
    2. Gu X
    3. Wang Y

    (2009) Color discrimination enhancement for dichromats using self-organizing color transformation

    Information Sciences 179:830–843.

    https://doi.org/10.1016/j.ins.2008.11.010

    • Google Scholar
34. 1. Machado GM
    2. Oliveira MM
    3. Fernandes LAF

    (2009) A physiologically-based model for simulation of color vision deficiency

    IEEE Transactions on Visualization and Computer Graphics 15:1291–1298.

    https://doi.org/10.1109/TVCG.2009.113

    • PubMed
    • Google Scholar
35. 1. Machado GM
    2. Oliveira MM

    (2010) Real‐time temporal‐coherent color contrast enhancement for dichromats

    Computer Graphics Forum 29:933–942.

    https://doi.org/10.1111/j.1467-8659.2009.01701.x

    • Google Scholar
36. Software

    1. Maechler M

    (2024) Diptest: hartigan’s dip test statistic for unimodality - corrected, version 0.77-1

    CRAN.

    https://cran.r-project.org/web/packages/diptest/diptest.pdf
37. 1. Nakauchi S
    2. Onouchi T

    (2008) Detection and modification of confusing color combinations for red‐green dichromats to achieve a color universal design

    Color Research & Application 33:203–211.

    https://doi.org/10.1002/col.20404

    • Google Scholar
38. 1. Nathans J
    2. Thomas D
    3. Hogness DS

    (1986) Molecular genetics of human color vision: the genes encoding blue, green, and red pigments

    Science 232:193–202.

    https://doi.org/10.1126/science.2937147

    • PubMed
    • Google Scholar
39. 1. Nelder JA
    2. Wedderburn RWM

    (1972) Generalized linear models

    Journal of the Royal Statistical Society. Series A (General) 135:370–384.

    https://doi.org/10.2307/2344614

    • Google Scholar
40. 1. Olson JM
    2. Brewer CA

    (1997) An evaluation of color selections to accommodate map users with color-vision impairments

    Annals of the Association of American Geographers 87:103–134.

    https://doi.org/10.1111/0004-5608.00043

    • Google Scholar
41. Software

    1. Ooms J

    (2021) Magick: advanced graphics and image-processing in R

    CRAN.

    https://docs.ropensci.org/magick/articles/intro.html
42. Software

    1. OpenJS Foundation

    (2024) Node.js, version v20.17.0

    OpenJS Foundation.

    https://nodejs.org/en
43. Software

    1. Ou J

    (2021) ColorBlindness: safe color set for color blindness, version 0.1.9

    R package version.

    https://CRAN.R-project.org/package=colorBlindness
44. 1. Pedregosa F

    (2011)

    Scikit-learn: machine learning in python

    Journal of Machine Learning Research 12:2825–2830.

    • Google Scholar
45. Software

    1. Piccolo S

    (2024) Colorblind image tester, version swh:1:rev:1f7c55070c5f36ae864171d3ceb58323489458d9

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:3b927e58ae4cd0c0cc712a634dab2f091a090a35;origin=https://github.com/srp33/colorblind_image_tester;visit=swh:1:snp:4584b11e238b1c07956a2bb91db32f2e53cb0884;anchor=swh:1:rev:1f7c55070c5f36ae864171d3ceb58323489458d9
46. Software

    1. Piccolo S
    2. Stevens HP

    (2024) Analysis code for “identifying images in the biology literature that are problematic for people with a color-vision deficiency.”, version swh:1:rev:86ebcead7bb37b7c89020eab1d5e1d948b307efe

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:47ac81a14ce7a965d4dda6f5de5272dcac5b044f;origin=https://github.com/srp33/bio_image_colorblindness;visit=swh:1:snp:09c5f311e1b54c55f32bd75a9290047e8a56a131;anchor=swh:1:rev:86ebcead7bb37b7c89020eab1d5e1d948b307efe
47. Book

    1. PMC Open Access Subset

    (2003)

    Bethesda (MD): national library of medicine

    PubMed Central (PMC).

    • Google Scholar
48. 1. Rasche K
    2. Geist R
    3. Westall J

    (2005) Re‐coloring images for gamuts of lower dimension

    Computer Graphics Forum 24:423–432.

    https://doi.org/10.1111/j.1467-8659.2005.00867.x

    • Google Scholar
49. Software

    1. R Development Core Team

    (2022) R: a language and environment for statistical computing, version 2.6.2

    R Foundation for Statistical Computing, Vienna, Austria.

    https://www.R-project.org/
50. 1. Ribeiro M
    2. Gomes AJP

    (2020) Recoloring algorithms for colorblind people

    ACM Computing Surveys 52:1–37.

    https://doi.org/10.1145/3329118

    • Google Scholar
51. 1. Robin X
    2. Turck N
    3. Hainard A
    4. Tiberti N
    5. Lisacek F
    6. Sanchez JC
    7. Müller M

    (2011) pROC: an open-source package for R and S+ to analyze and compare ROC curves

    BMC Bioinformatics 12:77.

    https://doi.org/10.1186/1471-2105-12-77

    • PubMed
    • Google Scholar
52. Conference

    1. Ruminski J
    2. Wtorek J
    3. Ruminska J
    4. Kaczmarek M
    5. Bujnowski A
    6. Kocejko T
    7. Polinski A

    (2010) Color transformation methods for dichromats

    in 3rd International Conference on Human System Interaction. pp. 634–641.

    https://doi.org/10.1109/HSI.2010.5514503

    • Google Scholar
53. Preprint

    1. Saladi SM
    2. Maggiolo AO

    (2019) JetFighter: towards figure accuracy and accessibility

    bioRxiv.

    https://doi.org/10.1101/2023.11.29.569308

    • Google Scholar
54. Preprint

    1. Sandler M
    2. Howard A
    3. Zhu M
    4. Zhmoginov A
    5. Chen LC

    (2019) MobileNetV2: inverted residuals and linear bottlenecks

    arXiv.

    https://arxiv.org/abs/1801.04381

    • Google Scholar
55. 1. Simunovic MP

    (2010) Colour vision deficiency

    Eye 24:747–755.

    https://doi.org/10.1038/eye.2009.251

    • PubMed
    • Google Scholar
56. 1. Stauffer R
    2. Mayr GJ
    3. Dabernig M
    4. Zeileis A

    (2015) Somewhere over the rainbow: how to make effective use of colors in meteorological visualizations

    Bulletin of the American Meteorological Society 96:203–216.

    https://doi.org/10.1175/BAMS-D-13-00155.1

    • Google Scholar
57. Book

    1. Still M

    (2006)

    The Definitive Guide to ImageMagick

    Apress.

    • Google Scholar
58. 1. Stockman A
    2. Sharpe LT

    (2000) The spectral sensitivities of the middle- and long-wavelength-sensitive cones derived from measurements in observers of known genotype

    Vision Research 40:1711–1737.

    https://doi.org/10.1016/s0042-6989(00)00021-3

    • PubMed
    • Google Scholar
59. 1. Swets JA

    (1988) Measuring the accuracy of diagnostic systems

    Science 240:1285–1293.

    https://doi.org/10.1126/science.3287615

    • PubMed
    • Google Scholar
60. 1. Tanner WP
    2. Swets JA

    (2001) A decision-making theory of visual detection

    Psychological Review 61:401–409.

    https://doi.org/10.1037/h0058700

    • Google Scholar
61. Conference

    1. Tanuwidjaja E

    (2014) Chroma: a wearable augmented-reality solution for color blindness

    Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Association for Computing Machinery. pp. 799–810.

    https://doi.org/10.1145/2632048.2632091

    • Google Scholar
62. Software

    1. TensorFlow

    (2019) TensorFlow.js machine learning for javascript developers

    TensorFlow.

    https://www.tensorflow.org/js
63. 1. Tsekouras GE
    2. Rigos A
    3. Chatzistamatis S
    4. Tsimikas J
    5. Kotis K
    6. Caridakis G
    7. Anagnostopoulos CN

    (2021) A novel approach to image recoloring for color vision deficiency

    Sensors 21:2740.

    https://doi.org/10.3390/s21082740

    • PubMed
    • Google Scholar
64. 1. Vinot F
    2. Brettel H
    3. Mollon JD

    (1999) Digital video colourmaps for checking the legibility of displays by dichromats

    Color Research & Application 24:243–252.

    https://doi.org/10.1002/(SICI)1520-6378(199908)24:4<243::AID-COL5>3.0.CO;2-3

    • Google Scholar
65. 1. Wang X
    2. Zhu Z
    3. Chen X
    4. Go K
    5. Toyoura M
    6. Mao X

    (2021) Fast contrast and naturalness preserving image recolouring for dichromats

    Computers & Graphics 98:19–28.

    https://doi.org/10.1016/j.cag.2021.04.027

    • Google Scholar
66. 1. Waskom MLS

    (2021) seaborn: statistical data visualization

    Journal of Open Source Software 6:3021.

    https://doi.org/10.21105/joss.03021

    • Google Scholar
67. 1. Wickham H
    2. Averick M
    3. Bryan J
    4. Chang W
    5. McGowan L
    6. François R
    7. Grolemund G
    8. Hayes A
    9. Henry L
    10. Hester J
    11. Kuhn M
    12. Pedersen T
    13. Miller E
    14. Bache S
    15. Müller K
    16. Ooms J
    17. Robinson D
    18. Seidel D
    19. Spinu V
    20. Takahashi K
    21. Vaughan D
    22. Wilke C
    23. Woo K
    24. Yutani H

    (2019) Welcome to the Tidyverse

    Journal of Open Source Software 4:1686.

    https://doi.org/10.21105/joss.01686

    • Google Scholar
68. Software

    1. Wickham H
    2. Hester J
    3. Ooms J

    (2021) XML 2: XML parse, version 1.3.6

    CRAN.

    https://cran.r-project.org/web/packages/xml2/index.html
69. Software

    1. Wilke C

    (2023) Colorblindr, version 1.4-0

    R-Forge.

    http://R-Forge.R-project.org
70. Conference

    1. Wong SC
    2. Gatt A
    3. Stamatescu V
    4. McDonnell MD

    (2016) Understanding data augmentation for classification: when to warp?

    2016 International Conference on Digital Image Computing. pp. 1–6.

    https://doi.org/10.1109/DICTA.2016.7797091

    • Google Scholar
71. Book

    1. Xie Y

    (2014) Knitr: A comprehensive tool for reproducible research in R

    In: Stodden V, Leisch F, Peng RD, editors. Implementing Reproducible Computational Research. Chapman and Hall/CRC. pp. 3–32.

    https://doi.org/10.1201/9781315373461-1

    • Google Scholar
72. 1. Zhu Z
    2. Toyoura M
    3. Go K
    4. Fujishiro I
    5. Kashiwagi K
    6. Mao X

    (2019a) Naturalness- and information-preserving image recoloring for red–green dichromats

    Signal Processing 76:68–80.

    https://doi.org/10.1016/j.image.2019.04.004

    • Google Scholar
73. 1. Zhu Z
    2. Toyoura M
    3. Go K
    4. Fujishiro I
    5. Kashiwagi K
    6. Mao X

    (2019b) Processing images for red–green dichromats compensation via naturalness and information-preservation considered recoloring

    The Visual Computer 35:1053–1066.

    https://doi.org/10.1007/s00371-019-01689-4

    • Google Scholar
74. 1. Zhu Z
    2. Mao X

    (2021) Image recoloring for color vision deficiency compensation: A survey

    The Visual Computer 37:2999–3018.

    https://doi.org/10.1007/s00371-021-02240-0

    • Google Scholar

Author details

1. Harlan P Stevens

   Department of Biology, Brigham Young University, Provo, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Validation, Investigation, Methodology, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

2. Carly V Winegar

   Department of Biology, Brigham Young University, Provo, United States

   Contribution

   Conceptualization, Software, Investigation, Methodology, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

3. Arwen F Oakley

   Department of Biology, Brigham Young University, Provo, United States

   Contribution

   Data curation, Formal analysis, Writing – review and editing

   Competing interests

   No competing interests declared

4. Stephen R Piccolo

   Department of Biology, Brigham Young University, Provo, United States

   Contribution

   Conceptualization, Resources, Software, Formal analysis, Supervision, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   stephen_piccolo@byu.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-2001-5640

• 385

  views

• 27

  downloads

• 0

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.