Introduction/Background The quality of Visual Inspection with Acetic acid (VIA) in screening of cervical cancer in many LMIC’s is often compromised by subjectivity and lack of personnel. Introducing an artificial intelligence decision support system (AI-DSS) could improve consistency of VIA assessment and could eventually enable task shifting with the potential of increasing coverage in low- and middle-income countries (LMICs). We introduced AI-DSS in different rural settings via a stepwise approach and assessed its field performance.

Methodology Cervical images were captured by providers in Bangladesh, Uganda and India during regular screening programs. Assessment of these images was performed by providers and the AI-DSS which could function off-line and on-the-spot. Assessment was followed by two independent expert panels of each 6 persons (one for Indian and one for Uganda/Bangladesh (UB) images). The consensus panel functioned as reference for the assessment of the images. Sensitivity, specificity, positive and negative predictive value were determined.

Results A total of 509 images were assessed by AI-DSS and experts. 36 (7%) images were excluded because of poor quality. Majority consensus (≥ 4 experts) was 89% (241/303) for the UB panel and 76% (157/204) for Indian panel. The images for which no majority consensus was reached were excluded from the analysis. The overall prevalence and accuracy were respectively 14.6% and 66.3%. Overall sensitivity and specificity for AI were 55.17% and 68.24% respectively. Positive and negative predictive values were 22.86% and 89.92% respectively.

Conclusion The accuracy of the AI-DSS is acceptable in line with most studies applied in LMICs. Therefore it is considered as a proper screening tool because of the high negative predictive value, indicating few false negative cases. Further adjusting to high-resolution cameras in new devices, improvement in quality of images over time and update of the algorithm will influence performance of AI-DSS in the future and will enable task shifting.

Disclosures Prevention and Screening Innovation Project – Towards Elimination of Cervical Cancer (PRESCRIP-TEC) is a research consortium project delivered through a collaboration of 15 consortium members. This project has received funding from the European Union’s Horizon 2020 research and innovation program grant agreement No 964270 and from the Ministry of Science and Technology, Department of Biomedical Technology in India, grant No 13213, under the Global Alliance for Chronic Diseases. International Agency for Research on Cancer (IARC), Leiden University Medical Centre (LUMC), and Uganda Cancer Institute (UCI) for availing the images used. Manipal Academy for Higher Education (MAHE) in India for availing the algorithm. Marconi laboratory in Makerere University, Uganda for providing the online tool. All the healthcare workers and experts for their time and voluntary effort for this study.