Using artificial intelligence on dermatology conditions in Uganda: A case for diversity in training data sets for machine learning
Open accessThe DOI matches its Crossref record; 31 of the 31 references it cites were found in a registry or an index.
Abstract
Introduction: Artificial intelligence (AI) in healthcare has gained momentum with advances in affordable technology that has potential to help in diagnostics, predictive healthcare and personalized medicine. In pursuit of applying universal non-biased AI in healthcare, it is essential that data from different settings (gender, age and ethnicity) is represented. We present findings from beta-testing an AI-powered dermatological algorithm called Skin Image Search, by online dermatology company First Derm on Fitzpatrick 6 skin type (dark skin) dermatological conditions. Methods: 123 dermatological images selected from a total of 173 images retrospectively extracted from the electronic database of a Ugandan telehealth company, The Medical Concierge Group (TMCG) after getting their consent. Details of age, gender and dermatological clinical diagnosis were analyzed using R on R studio software to assess the diagnostic accuracy of the AI app along disease diagnosis and body part. Predictability levels of the AI app was graded on a scale of 0 to 5, where 0-no prediction made and 1-5 demonstrating reducing correct prediction. Results: 76 (62%) of the dermatological images were from females and 47 (38%) from males. The 5 most reported body parts were; genitals (20%), trunk (20%), lower limb (14.6%), face (12%) and upper limb (12%) with the AI app predicting a diagnosis in 62% of image body parts uploaded. Overall diagnostic accuracy of the AI app was low at 17% (21 out of 123 predictable images) with varying predictability levels correctness i.e. 1-8.9%, 2-2.4%, 3-2.4%, 4-1.6%, 5-1.6% with performance along individual diagnosis highest with dermatitis (80%). Conclusion: There is a need for diversity in the image datasets used when training dermatology algorithms for AI applications in clinical decision support as a means to increase accuracy and thus offer correct treatment across skin types and geographies.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures





Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“AIs trained primarily on lighter skin tones may not generalize as well when tested on darker skin157 .”
From Deep learning-enabled medical computer vision (2021) · cites this paper as [157]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hay, R. J., Johns, N. E., Williams, H. C., Bolliger, I. W., Dellavalle, R. P., Margolis, D. J., Marks, R., Naldi, L., Weinstock, M. A., Wulf, S. K., et al. (2014) · Journal of Investigative Dermatology · doi:10.1038/jid.2013.446 · Cited by 1,706
Potential Biases in Machine Learning Algorithms Using Electronic Health Record Data
Gianfrancesco, M. A., Tamang, S., Yazdany, J., & Schmajuk, G. (2018) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2018.3763 · Cited by 1,395
Findings from the Global Burden of Disease Study 2021
Murray, C. J. L. (2024) · The Lancet · doi:10.1016/s0140-6736(24)00769-4 · Cited by 1,057
A comprehensive review on automation in agriculture using artificial intelligence
Jha, K., Doshi, A., Patel, P., & Shah, M. (2019) · Artificial Intelligence in Agriculture · doi:10.1016/j.aiia.2019.05.004 · Cited by 1,005
Feasey, N., Wansbrough-Jones, M., Mabey, D. C. W., & Solomon, A. W. (2010) · British Medical Bulletin · doi:10.1093/bmb/ldp046 · Cited by 817
Letting Artificial Intelligence in Education Out of the Box: Educational Cobots and Smart Classrooms
Timms, M. J. (2016) · International Journal of Artificial Intelligence in Education · doi:10.1007/s40593-016-0095-y · Cited by 428
Global Burden of Skin Disease: Inequities and Innovations
Seth, D., Cheldize, K., Brown, D., & Freeman, E. E. (2017) · Current Dermatology Reports · doi:10.1007/s13671-017-0192-7 · Cited by 366
The Ugly Truth About Ourselves and Our Robot Creations: The Problem of Bias and Social Inequity
Howard, A., & Borenstein, J. (2018) · Science and Engineering Ethics · doi:10.1007/s11948-017-9975-2 · Cited by 337
Marchetti, M. A., Codella, N. C. F., Dusza, S. W., Gutman, D. A., Helba, B., Kalloo, A., Mishra, N., Carrera, C., Celebi, M. E., DeFazio, J. L., et al. (2018) · Journal of the American Academy of Dermatology · doi:10.1016/j.jaad.2017.08.016 · Cited by 330
Artificial Intelligence in Pathology
Chang, H. Y., Jung, C. K., Woo, J. I., Lee, S., Cho, J., Kim, S. W., & Kwak, T.-Y. (2019) · Journal of Pathology and Translational Medicine · doi:10.4132/jptm.2018.12.16 · Cited by 246
Cited by
Items 101–111 of 111
Xu, G., Wu, Y., Jia, Z., Hu, J., & Shi, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-05870-6_17 · Cited by 3
Designing Nutritional Interventions for Mental Health
Hendi, N. I., Abouelmagd, M. E., Sharkawy, A. M., Jasim, E. M., & Amin, A. M. (2025) · Nutritional neurosciences · doi:10.1007/978-981-95-1983-5_2
Coates, S. J., Yang, F., Hill, C., Xue, Z., Rajaraman, S., Semeere, A., Ayanga, R., Laker‐Oketta, M., Byakwaga, H., Lukande, R. L., et al. (2026) · JCO Global Oncology · doi:10.1200/go-25-00236 · Cited by 4
AI and automation in assisted reproduction
Lorimer, J., McLachlan, R., Zander‐Fox, D., O’Bryan, M. K., & Nosrati, R. (2026) · Nature Reviews Bioengineering · doi:10.1038/s44222-026-00454-2 · Cited by 1
“AI models trained on narrow demographic samples often fail to generalize 171,174,176 .”
Toward personalised cancer care: AI in precision and molecular oncology
Sinha, A., Anzar, D., Owusu-Boampong, A. B., Nkrumah-Boateng, P. A., Mannan, K. M., Ali, S. H., Roy, S., & Wireko, A. A. (2026) · European journal of medical research · doi:10.1186/s40001-026-04670-9
“However, these algorithms are often trained predominantly on Caucasian patients and may be less effective at detecting skin lesions in darker skin tones, with one application having significantly lower diagnostic accuracy on Fitzpatrick 6 skin type [170, 171].”
Mulyowa, G. K., Engwau, T., Galiwango, M., Kamuganga, F., Mirembe, S. K., Onguti, A. G., Mwavu, R., Wasswa, W., Kaggwa, F., & Obua, C. (2026) · Clinical Cosmetic and Investigational Dermatology · doi:10.2147/ccid.s603936
Opting In? Information About AI Training Affects Human Behavior
Treiman, L. S., Ho, C.-J., & Kool, W. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812384 · Cited by 3
“Kamulegeya et al. [38] found that an algorithm trained on a dataset of predominantly light-skinned patients performed 50% less accurately on darker skin tones.”
Parsons, S. P. (2026) · International Journal of Public Sector Management · doi:10.1108/ijpsm-08-2025-0356
A Human-Computer Interaction Approach to Medical Explainable AI (XAI)
Schwerk, A. M. K., & Blum, L. (2026) · Springer eBooks · doi:10.1007/978-981-95-3811-9_5 · Cited by 1
Pérez-Lalinde, S., & Flores, F. (2026) · Skin Research and Technology · doi:10.1111/srt.70385 · Cited by 1
Li, Z., Chen, K., Ji, Y., Wu, H., & Lu, Q. (2026) · Skin · doi:10.2738/skin.2026.0011
Metadata
- Authors
- Kamulegeya, L. H., Okello, M., Bwanika, J. M., Musinguzi, D., Lubega, W., Rusoke, D., Nassiwa, F., & Börve, A.
- Affiliations
- Sahlgrenska University Hospital (SE)
- Corresponding author
- Kamulegeya, Louis Henry
- Venue
- African Health Sciences
- Publisher
- openRxiv
- Published
- 2019-10-31
- Volume
- 23
- Issue
- 2
- Pages
- 753-63
- ISSN
- 1680-6905 (ISSN-L)
- PMC
- PMC10782289
- OpenAlex
- W4387584778
- Cited by
- 111
- References
- 30
- Citations by year
- 2026: 15 · 2025: 32 · 2024: 26 · 2023: 18 · 2022: 10 · 2021: 7 · 2020: 3
- FWCI
- 3.63
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- cc-by
- Locations
- African Health Sciences (publishedVersion, cc-by)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Oncology › Cutaneous Melanoma Detection and Management
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- dermatology, artificial intelligence, Fitzpatrick skin type, dark skin, training data diversity, algorithmic bias, diagnostic accuracy, telemedicine, Uganda, skin inflammation
- MeSH terms
- Machine Learning, Artificial Intelligence, Dermatology, Female, Humans, Male, Retrospective Studies, Uganda
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.