Skip to main content
Fig. 5 | Infectious Agents and Cancer

Fig. 5

From: Validation in Zambia of a cervical screening strategy including HPV genotyping and artificial intelligence (AI)-based automated visual evaluation

Fig. 5

Precancer risk stratification among low-prevalence HIV-positive Zambia study population by HPV and AVE combined results. This figure is an extension of Fig. 4 extending the risk discrimination to demonstrate the intended use case of PAVE, triaging HPV-positive individuals with HPV genotyping and AVE. The population is first tested for HIV, and this figure shows the risk discrimination among HIV-positive patients first tested with HPV and then with AVE. As demonstrated in previous images, 15% precancer/cc risk of HIV-positive patients will vary between 48 and 1.4% (HPV 16+ and HPV HR-negative, respectively) after being tested for HPV genotype. If we apply AVE test after HPV genotype, we can observe even finer risk discrimination such that 48% precancer/cc risk of HIV-positive and HPV 16+ patients will increase to 72% for AVE result precancer/cc and decrease to 21% for AVE result normal. The highest risk group is HPV 16+ and AVE precancer/cc, followed by other HPV-positive groups and AVE precancer/cc. The lowest risk groups are HPV HR-negative and AVE normal/indeterminate with 0–2.0% precancer/cc risk. 37 of HIV+ individuals do not have any J8 images (their images were captured by other camera types), so they are not included in this analysis. *No cases observed in these categories

Back to article page