Project summary

Cervical cancer remains a major public health challenge in Uganda, where it is the leading cause of cancer-related deaths among women, with approximately 6,959 new cases and 4,607 deaths annually. The national age-standardized incidence rate of 56.2 per 100,000 women is among the highest globally, and over 80% of cases are diagnosed at advanced stages due to limited access to timely and accurate diagnostics. Histopathology is the diagnostic gold standard; however, Uganda faces a severe shortage of pathologists fewer than 30 nationwide resulting in diagnostic delays, variability, and inequitable access, particularly for rural populations.

This study aims to design and validate a machine learning (ML)–based digital histopathology platform for cervical cancer diagnosis in Southwestern Uganda. Using approximately 1,100 retrospectively archived cervical biopsy slides (2018–2024) from Mbarara Regional Referral Hospital and Mbarara University of Science and Technology, whole-slide images will be digitized and analyzed using deep learning models trained on locally sourced data. The platform will integrate ML predictions with a secure, user-friendly digital pathology interface and a human-in-the-loop framework to support expert oversight.

Diagnostic accuracy (sensitivity, specificity, predictive values, and AUC) of the ML platform will be evaluated against consensus diagnoses from expert pathologists. By generating locally validated evidence, this study seeks to improve diagnostic efficiency, consistency, and access, supporting earlier detection and aligning with Uganda’s National Cancer Control Plan and the WHO Global Strategy to Eliminate Cervical Cancer.

 

Lead Principal Investigator: Dr Atwine Raymond

 

Co- Principal Investigator: Dr Atwine Daniel

 

Implementers: Collaborative project between Mbarara University of Science and Technology and  SRF Research and Training Centres, Mbarara, Uganda

 

Funding Source: MUST Internal Research funds-DRGT

 

Current status: Ongoing regulatory approvals

 

Duration: 10 months

 

Start date: October 2025

 

Expected End date: June 2026