The Fetal Age Machine learning Initiative (FAMLI) is a collaboration between UNC GWH, NC State University, and the University of Zambia School of Medicine. It is funded by the Bill and Melinda Gates Foundation.
Our overarching goal of FAMLI is to develop a robust, inexpensive, widely-deployable ultrasound device that can assess gestational age and other important obstetric conditions while requiring minimal operator expertise. The project will produce a suite of critical resources – including data, new sensor technology, and machine learning techniques – that will be made widely available to computer vision / AI groups working in pursuit of our overarching goal. We will leverage the obstetrical sonography resources of the UNC Department of Obstetrics and Gynecology which has access to diverse patient populations in both North Carolina and Zambia (through a longstanding partnership with the University of Zambia). We will produce large sets of ultrasound data that can be used to train machine learning algorithms to assess gestational age and make other key diagnoses. With colleagues at NC State University we will engineer new sensor technologies that can be deployed alongside the ultrasound probe to facilitate computer-assisted image interpretation.
Figure: Blind sweep “cineloop” (midline; inferior to superior) of a 20 week fetus. Many structures important to fetal biometry come into focus in this single sweep.