@article{103571,
  keywords = {Blindness, machine learning, R-workflow, Surveillance, Trachoma},
  author = {Kulohoma B and Wesonga C},
  title = {Scaling trachoma surveillance in endemic areas using machine learning},
  abstract = {<p><strong>Background</strong></p>

<p>Trachoma remains a leading infectious cause of blindness in endemic regions despite progress towards global elimination. Accurate and scalable diagnosis remains challenging in areas with limited ophthalmic expertise. We developed and evaluated a reproducible transfer learning model using eyelid image-based data.</p>

<p><strong>Methods</strong></p>

<p>We retrospectively analysed anonymised inner eyelid photographs collected from trachoma prevalence surveys conducted in Ethiopia, Tanzania, Australia, Solomon Islands, Colombia, the Gambia, and Guatemala (<em>n</em> = 572 images). Images were categorized as trachoma (<em>n</em> = 251) or not trachoma (<em>n</em> = 321) based on consensus grades from the Global Trachoma Mapping Project (GMTP) certified graders. Data were processed and analysed in R (version 4.4.1) using the keras and tensorflow packages interfaced with Python (version 3.10) through reticulate. A ResNet50 convolutional neural network pretrained on ImageNet was fine-tuned for binary classification. The model was trained for up to 30 epochs with early stopping and adaptive learning-rate reduction. Performance on a held-out validation set (<em>n</em> = 62, 12%) was evaluated using accuracy, sensitivity, specificity, and Cohen’s κ.</p>

<p><strong>Results</strong></p>

<p>The ResNet50 model achieved an overall accuracy of 85.4% (95% CI 76.3–92%) on the validation set. The model achieved a sensitivity of 80.9% for detecting trachoma and a specificity of 90.5% for correctly identifying non-trachoma eyes. Agreement between predictions and grader labels was substantial (κ = 0.71, 95% CI 0.58–0.84).</p>

<p><strong>Conclusion</strong></p>

<p>Our findings demonstrate the feasibility of using a reproducible R-based deep learning pipeline for automated trachoma classification in large-scale surveys. These findings demonstrate the feasibility of automated trachoma classification using a transfer-learning framework. Larger datasets will be required before operational deployment in surveillance programmes.</p>
},
  year = {2026},
  journal = {BMC infectious diseases},
  pages = {1 - 12},
  month = {04/2026},
  issn = {1471-2334},
  url = {https://link.springer.com/content/pdf/10.1186/s12879-026-13215-8_reference.pdf},
  doi = {10.1186/s12879-026-13215-8},
  language = {ENG},
}