CBMI Active Projects 

Knowledge of APOL1 status early in pregnancy may be useful for preeclampsia prediction and pregnancy risk stratification, in order to reduce maternal and perinatal mortality associated with hypertensive disorders of pregnancy. Our study’s overarching goal is to better quantify the relationship between fetal (and maternal) APOL1 status and preeclampsia, and ultimately to develop better low-cost methods to predict high-risk pregnancies and improve maternal and infant health outcomes.

Global disease surveillance is fragmented across diseases, countries, funding sources, and a wide range of dynamic heterogeneous clinical and laboratory data sources. The SIEMA (Semantics, Interoperability, and Evolution for Malaria Analytics) project is implementing advanced technologies to support data, language and semantic interoperability and to help integrate dynamic surveillance data across multiple scales, in provision of transparent and scalable tools for decision making for malaria control and elimination. Our development efforts will be focused on sentinel sites in selected African countries, including Uganda and Burkina Faso (to ensure language interoperability between French and English systems). The pilot project has already demonstrated the feasibility and value of our method and tool for malaria surveillance community.

PopHR platform to semantically and statistically investigate the relationships between dwelling conditions and health outcomes, and merging multiple data sources to unpack disproportionate health outcome distributions, such as asthma and obesity prevalence in Memphis, TN.

Real-Time Event Stream Analytics (RESA) for Prediction of Pediatric Multi-Organ Dysfunction Syndrome (P-MODS) has been developed through several prior work on neonatal apnea, sepsis, and physiological deterioration, demonstrating the viability of a stable and reliable real-time analytic environment. Our central hypothesis is that by applying real time analysis of physiologic data streams, we will be able to accurately predict P-MODS earlier and reduce time to intervention and treatment for critically ill children.

Detection of cardiac abnormalities from raw ECG signals

Analysis of unstructured EMR data for surgical risk stratification

Understanding the sources of racial disparities in surgery outcome