Utility of a machine-guided tool for assessing risk behaviour associated with contracting HIV in three sites in South Africa.

Introduction: Digital data collection and the associated mobile health technologies have allowed for the recent exploration of artificial intelligence as a tool for combatting the HIV epidemic. Machine learning has been found to be useful both in HIV risk prediction and as a decision support tool for guiding pre-exposure prophylaxis (PrEP) treatment. This paper reports data from two sequential studies evaluating the viability of using machine learning to predict the susceptibility of adults to HIV infection using responses from a digital survey deployed in a high burden, low-resource setting. Methods: 1036 and 593 participants were recruited across two trials. The first trial was a cross-sectional study in one location and the second trial was a cohort study across three trial sites. The data from the studies were merged, partitioned using standard techniques, and then used to train and evaluate multiple different machine learning models and select and evaluate a final model. Variable importance estimates were calculated using the PIMP and SHAP methodologies. Results: Characteristics associated with HIV were consistent across both studies. Overall, HIV positive patients had a higher median age (34 [IQR: 29-39] vs 26 [IQR 22-33], p < 0.001), and were more likely to be female (155/703 [22%] vs 107/927 [12%], p < 0.001). HIV positive participants also had more commonly gone a year or more since their last HIV test (183/262 [70%] vs 540/1368 [39%], p < 0.001) and were less likely to report consistent condom usage (113/262 [43%] vs 758/1368 [55%], p < 0.001). Patients who reported TB symptoms were more likely to be HIV positive. The trained models had accuracy values (AUROCs) ranging from 78.5% to 82.8%. A boosted tree model performed best with a sensitivity of 84% (95% CI 72-92), specificity of 71% (95% CI 67-76), and a negative predictive value of 95% (95% CI 93-96) in a hold-out dataset. Age, duration since last HIV test, and number of male sexual partners were consistently three of the four most important variables across both variable importance estimates. Conclusions: This study has highlighted the synergies present between mobile health and machine learning in HIV. It has been demonstrated that a viable ML model can be built using digital survey data from an low-middle income setting with potential utility in directing health resources.

Hepatitis B and C virus--prevalence and prevention in health care workers.

BACKGROUND: Hepatitis B and C viruses are important causes of liver related morbidity and mortality. We aimed at determining the presence of hepatitis B and C virus infections in the health care workers (HCWs) and their compliance for the HBV vaccination. METHODS: Three thousand five hundred and fifty six health care workers were screened for HBsAg and 115 for anti-HCV by ELISA. HBsAg negative individual were offered HBV vaccination and record of their compliance was kept. Anti-HBs titers were determined one month after 2nd or 3rd dose of vaccine in 273 subjects. RESULTS: Out of 3556 health care workers, 61 (1.7%) were found to be positive for HBsAg. One out of 115 HCWs (0.87%) was found to be positive for anti-HCV. Fifteen percent of HCWs received only one dose, 26% received two doses 59% received three doses and 2.5% also received the booster dose of the HBV vaccine. All those tested had anti-HBs titers more than 10 mUI/ml. CONCLUSION: In HCWs, HBsAg and anti-HCV prevalence was found to be 1.7% and 0.87% respectively. HCWs in our hospital, despite the awareness on HBV and HCV infection are noncompliant for HBV vaccination.

Percutaneous delivery of a heparin-impregnated collagen stent-graft in a porcine model of atherosclerotic disease.

RATIONALE AND OBJECTIVES: To develop collagen stent-grafts impregnated with heparin to improve the biocompatibility of endovascular stents and to design a percutaneous delivery system for graft deployment in a swine model. METHODS: Heparin-impregnated collagen stent-grafts were deployed, and follow-up angiograms were obtained every 15 minutes for 90 minutes to assess acute thromboses and again at 2 and 4 weeks afterward to assess patency. If stenosis or occlusion was detected at the 2-week evaluation, guidewire passage across the lesion was attempted and angioplasty was performed. If stenosis or occlusion was present at the 4-week evaluation, only guidewire passage was attempted; thereafter, the animals were killed and the stent-grafts were harvested and reviewed by a vascular pathologist. RESULTS: Group A represents a feasibility study to optimize the deployment method applied in groups B and C. Fifteen of 17 stent-grafts were successfully deployed using this method. In group B, 89% of grafts were successfully deployed; 12% were patent at 2 weeks and none at 4 weeks. In group C, a 10-minute inflation time was added to the deployment procedure; 88% of grafts were successfully deployed and 28% were patent at 2 weeks and 14% at 4 weeks. Extensive luminal thrombosis and myointimal hyperplasia were present in every case. CONCLUSIONS: A method was developed for percutaneous implantation of collagen stent-grafts into peripheral vessels. The heparin-impregnated grafts did not prevent vessel restenosis. Modification of the graft-processing technique may improve patency.

Dialysis access graft thrombolysis: randomized study of pulse-spray versus continuous urokinase infusion.

PURPOSE: To compare pulse-spray to continuous-infusion thrombolysis with high-dose urokinase in thrombosed dialysis access grafts. METHODS: A prospective randomized controlled trial was performed. From August 1992 to September 1993, 30 thrombosed polytetrafluoroethylene (PTFE) grafts in 24 patients were included, 15 grafts in each group. The success of thrombolysis, mean time to thrombolysis, mean urokinase dose, and 60-day patency rate were evaluated. RESULTS: In the pulse-spray group, the mean time to thrombolysis was 72 min with a mean urokinase dose of 560,000 U. The 60-day patency rate was 71%. In the continuous-infusion group, the mean infusion time to thrombolysis was 55 min with a mean dose of 479,000 U. The 60-day patency rate was 73%. CONCLUSION: No statistically significant difference was found between the two techniques in the mean time to thrombolysis, the mean urokinase dose used, or the 60-day patency rate.