REM - Regression models based on expectation maximization algorithm

Loc Nguyen, Thu-Hang Thi Ho

Product download.

Git source.

REM project implements regression models based on expectation maximization (EM) algorithms in case of missing data, which is used for fetal weight estimation. Especially, fetal weight estimation before delivery is important in obstetrics, which assists doctors diagnose abnormal or diseased cases. Linear regression based on ultrasound measures such as bi-parietal diameter (bpd), head circumference (hc), abdominal circumference (ac), and fetal length (fl) is common statistical method for weight estimation but the regression model requires that time points of collecting such measures must not be too far from last ultrasound scans.

Two methods which are proposed and implemented in REM project are based on expectation maximization (EM) algorithm so that ultrasound measures can be taken at any time points in gestational period and so early weight estimation is achieved. The first method is called dual regression expectation maximization (DREM) algorithm in which only fetal age is missed whereas ultrasound measures must be complete. Currently DREM was completed and published in the article “Early Fetal Weight Estimation with Expectation Maximization Algorithm” in Experimental Medicine by International Technology and Science Publications. The second method is called Regression Expectation Maximization (REM) algorithm to solve the hazard problem in which fetal weight, fetal ages, and ultrasound measures can be missing. Currently, REM was completed but it is not published yet.

Prof. Dr. Thu-Hang Thi Ho granted 1600USD budget for REM project. Duration of project is from March 2018 to September 2018. Members of REM project are Loc Nguyen and Thu-Hang Thi Ho.