AI tool helps identify patients at high risk for HIV, STIs
Click Here to Manage Email Alerts
Key takeaways:
- MySTIRisk, an AI-based tool, successfully estimated infection risk scores for HIV, syphilis, gonorrhea and chlamydia.
- Patients considered “high risk” had significantly higher positivity for all four infections.
Using MySTIRisk, an artificial intelligence-based risk assessment tool, researchers were successfully able to create infection risk scores to identify patients at higher risk for HIV, syphilis, gonorrhea and chlamydia.
“The Artificial Intelligence and Modelling in Epidemiology Program from the Melbourne Sexual Health Centre has developed a web and [artificial intelligence (AI)-based risk] assessment tool for HIV and sexually transmitted infections [called] MySTIRisk,” Phyu M. Latt, PhD candidate and research assistant at the Melbourne Sexual Health Center, told Healio.
“When displaying personalized risk scores to users, we found most participants preferred seeing clear categorization of whether they were at high or average risk for infections rather than just statistics. This prompted us to identify optimal risk score thresholds to effectively stratify individuals as high-risk vs. average-risk,” Latt said, adding that, “in practice, knowing one’s risk of HIV and STIs enables targeted and better HIV testing and prevention.”
The researchers conducted a retrospective cross-sectional study using data from 216,252 HIV, 227,995 syphilis, 262,599 gonorrhea, and 320,355 chlamydia consultations at the Melborne Health Centre between 2008 and 2022.
The team applied MySTIRisk machine learning models to estimate infection risk scores by assessing predictors such as gender, age, country of birth, men who reported having sex with men, presence of STI symptoms, number of partners, condom use, injection drug use, past STIs, contact with STI diagnoses and sexual partners outside Australia/New Zealand.
Through the study, researchers were able to identify optimal risk score thresholds categorizing patients as "high-risk" vs. “average-risk" for HIV, syphilis, gonorrhea and chlamydia. Specifically, the study determined that the HIV high-risk score cutoff was 0.56, with 86% sensitivity (95% CI, 82.9%-88.7%) and 65.6% specificity (95% CI, 65.4%-65.8%). Of participants accessed, 35% percent were classified as high risk, which accounted for 86% of HIV cases.
Lutt added that 0.7% of the high-risk group had HIV compared with just 0.06% of the average-risk group, demonstrating that the tool can “accurately identify patients most needing prevention and testing services based on their underlying level of risk.”
Additionally, the tool demonstrated that the corresponding cutoffs were 0.49 for syphilis with 77.6% sensitivity and 78.1% specificity, 0.52 for gonorrhea with 78.3% sensitivity and 71.9% specificity and 0.47 for chlamydia with 68.8% sensitivity and 63.7% specificity. According to the study, these high-risk groups accounted for 78% of syphilis, 78% of gonorrhea and 69% of chlamydia cases, with the odds of positivity being significantly higher in the high-risk group than otherwise across all infections (12.3 for syphilis, 9.2 for gonorrhea and 3.9 for chlamydia).
Based on these findings, Latt said that the AI-based MySTIRisk tool showed promise in leveraging both behavioral and clinical data to estimate individualized infection risk scores for HIV and other common STIs.
“Risk assessment tools leveraging complex real-world data have shown promise to guide resource prioritization through the identification of patients most needing targeted testing and prevention services,” Latt said. “If mindfully integrated under clinician guidance, machine learning approaches like MySTIRisk may help streamline sexual health services, though ultimately, decisions should use risk stratification to support rather than supplant sound provider judgment.”
Collapse
Read more about
Click Here to Manage Email Alerts