Project Summary / Abstract Preeclampsia is a complication affecting 2 – 8% of all pregnancies and results in significant maternal and neonatal morbidity and mortality. The invention to be researched in this grant detects preeclampsia and hypertension in patients with the phones that they already own. It improves outcomes by increasing access to care among underserved populations and monitoring frequency among high risk patients. Patients record themselves using their own phones which takes only a few minutes. The long term objective is to commercialize a product which enables any of 7 billion phones to assist in preeclampsia or gestational hypertension detection. The invention is intended to be used at home by patients via a phone app with results sent to the prescribing obstetrician (OB). The purpose is to alert the OB if immediate physical examination is needed and to initiate possible preeclampsia management protocol. In these situations, there are no other comparable testing options. The invention includes both improvements to clinical practice and to science. Current clinical practice standard of care requires patients to be examined in an OB office. The invention allows preeclampsia and hypertension to be detected via patient self-examination from locations outside the OB office. This improves upon current technology such as home blood pressure cuffs and in-office dipstick urinalysis. In terms of improvements to science, the invention consists of both a Universal Translator (UT) and Deductive Intelligence (DI). UT allows for any mobile phone to capture body acoustic data. DI is a novel physics based approach to creating classifier algorithms from passively received time series data, such as mobile phone recordings received from the UT. The invention analyzes hemodynamics and extracts pertinent physics based features from which a classifier algorithm is based. This will be the third large scale human study that demonstrates classification of cardio - pulmonary functionality. It follows a published study on COVID detection as well as a study under peer review on the ability to reproduce echocardiogram estimates of ejection fraction. The echocardiogram study establishes the protocol to be used in the proposed research. In a recent preliminary study, 46 pregnant women were recorded at the aortic site. The resulting model was able to determine which patients had high blood pressure and to accurately determine which patients had complications. These studies support our hypothesis that acoustic hemodynamic data captured by OEM phone microphones, at the aortic auscultation site and at the upper arm, can be used to identify patients who have preeclampsia and/or hypertension. The study has two primary aims. The first is to demonstrate proof of concept that the invention enables ordinary mobile phones to reproduce physicians’ diagnosis of pre-eclampsia. The second aim is to demonstrate proof of concept that the invention enables ordinary mobile phones to reproduce cuff measurements of blood pressure and to detect hypertension. The approach is based on recruitment from both outpatient clinics and inpatient hospital settings. Patients already labeled with preeclampsia or high blood pressure will be enrolled as well as control patients without disease conditions. Patients will be recorded by clinicians with a custom phone app. Algorithms will be developed based on these recordings using UT and DI. Results will be tested and analyzed using AUC, sensitivity/specificity, accuracy as well adj RSQ for the linear regression analysis. The team consists of a broad range of talent including the inventor of the technology, the Maternal Fetal Medicine Division Director at the hospital, the research director of emergency medicine for the hospital running the tests, statistical expertise, and project management.

The research supports development of technology that allows ordinary mobile phones to be used to detect preeclampsia and hypertension in patients remotely. The intent is to improve access to care as well as the ability to increase monitoring frequency. The technology is scientifically important in that it uses a novel hemodynamics based approach in creating classifier algorithms from passively received time series data.