Advancing Knowledge
Publications from our team of researchers and collaborators
Our Publications
A Pilot Study Protocol for AI-Assisted Interpretation of Chest X-rays for Pulmonary Abnormalities in Uganda
Authors: Obungoloch J, Tumusiime J, Nkwanga J, Godfrey MR, Mbusa C, Kaggwa F, Celi LA, Haberer JE, Wasswa W
Journal: Cureus (18(6), 2026)
Pages: e110504
Timely access to chest X-ray imaging and interpretation remains a practical challenge in routine pulmonary care in Uganda. This pilot protocol describes a cross-sectional study to develop a locally derived dataset of annotated CXR images linked with clinical metadata and evaluate the feasibility of machine learning models to support diagnosis of pulmonary conditions.
Clinical Features and Risk Factors of Dermatitis Cruris Pustulosa et Atrophicans in Southwestern Uganda: A Cross-Sectional Study
Authors: Mulyowa G, Engwau T, Galiwango M, Kamuganga F, Mirembe S, Aloyo G, Mwavu R, Wasswa W, Kaggwa F, Obua C
Journal: Clin Cosmet Investig Dermatol (19, 2026)
Pages: 1-9
DOI: 10.2147/CCID.S603936
Dermatitis cruris pustulosa et atrophicans (DCPA) is a marginally recognized chronic skin inflammatory condition in Uganda. This cross-sectional study of 405 participants in southwestern Uganda found DCPA mainly affects young females and identified oily cosmetic products as a contributory factor, highlighting the need for improved diagnostic tools and targeted interventions.
Application of AI to Ultrasonographic Images to Aid the Clinical Care of Pregnant Women With Pre-eclampsia in Uganda: A Protocol for a Pilot Study
Authors: Godfrey MR, Atwiine F, Atukunda EC, Celi LA, Mwavu R, Kaggwa F, Haberer JE, Wasswa W
Journal: Cureus (18(1), 2026)
Pages: e101406
Access to obstetric ultrasound and trained sonographers remains limited in Uganda despite the high burden of pre-eclampsia. This pilot protocol describes a cross-sectional study to create an annotated Doppler ultrasonographic image database linked to clinical metadata and develop machine learning models to predict maternal and fetal complications of pre-eclampsia.
Unmasking biases and navigating pitfalls in the ophthalmic artificial intelligence lifecycle: A narrative review
Authors: Nakayama LF, Matos J, Quion J, Novaes F, Mitchell WG, Mwavu R, Hung CJJ, Santiago APD, Phanphruk W, Cardoso JS, Celi LA
Journal: PLOS Digit Health (3(10), 2024)
Pages: e0000618
DOI: 10.1371/journal.pdig.0000618
Over the past 2 decades, exponential growth in data availability, computational power, and newly available modeling techniques has led to an expansion in interest, investment, and research in Artificial Intelligence (AI) applications. Ophthalmology is one of many fields that seek to benefit from AI given the advent of telemedicine screening programs and the use of ancillary imaging. However, before AI can be widely deployed, further work must be done to avoid the pitfalls within the AI lifecycle.
Assessment of Clinical Metadata on the Accuracy of Retinal Fundus Image Labels in Diabetic Retinopathy in Uganda: Case-Crossover Study Using the Multimodal Database of Retinal Images in Africa
Authors: Arunga S, Morley KE, Kwaga T, Morley MG, Nakayama LF, Mwavu R, Kaggwa F, Ssempiira J, Celi LA, Haberer JE, Obua C
Journal: JMIR Form Res (8, 2024)
Pages: e59914
DOI: 10.2196/59914
Labeling color fundus photos (CFP) is an important step in the development of artificial intelligence screening algorithms for the detection of diabetic retinopathy (DR). Most studies use the International Classification of Diabetic Retinopathy (ICDR) to assign labels to CFP, plus the presence or absence of macular edema (ME).
Ophthalmology Optical Coherence Tomography Databases for Artificial Intelligence Algorithm: A Review
Authors: Restrepo D, Quion JM, Do Carmo Novaes F, Azevedo Costa ID, Vasquez C, Bautista AN, Quiminiano E, Lim PA, Mwavu R, Celi LA, Nakayama LF
Journal: Semin Ophthalmol (39(3), 2024)
Pages: 200-210
DOI: 10.1080/08820538.2024.2312464
This review examines optical coherence tomography (OCT) databases used for AI algorithm development in ophthalmology, highlighting the importance of diverse datasets for accurate AI models.
Publication Impact
Chart visualization will be displayed here
Publication Timeline
Two publications in PLOS Digital Health
Publication in Seminars in Ophthalmology
Publication in Seminars in Ophthalmology
Publication in Lancet Digital Health
Publication in JMIR Formative Research
Publication in PLOS Digital Health
Publication in Cureus (Sono AI protocol)
Publications in Cureus and Clinical, Cosmetic and Investigational Dermatology