Predictive Assessment of Fetus Features Using Scanned Image Segmentation Techniques and Deep Learning Strategy

1. Introduction

One of the most important indicators of fetus survival is weight of the baby. Identifying early stage healthiness of the fetus will leads to the greater safety of the baby. The factors that play significant part in finding out the weight of the baby are Head Circumference, Bi-Parietal diameter, Femur Length and Abdominal Circumference. The main factors that influence the fetus health is fetus weight and this plays very important role in both fetus and mother’s health. The demographics of the fetus are very important as well however, the food taken during the pregnancy period and early key factors will regulate the weight of the fetus. In the current work, we have to measure the weight of the fetus by using advanced computer vision techniques. Fetus growth will be generally measure during the follow up scans of 11th week and 13th week of pregnancy in phase I and 18th to 20th in phase II. So the weight of fetus can be calculated only during phase II. However, with proposed method every week a scan has to be taken and the weight of the baby will be presented with the scanned report by which physician can measure the accurate fetus weight.

As a standard practice when we measure the scans the doctors check the height of the fetus and measure the weight as well. Incase if there not enough growth an immediate alert will be raised and preventive medicine will be given to the patients. The standard method helps in measuring the bump form your public bone to help and feel the hands on top of the womb and compare with what expected is the standard procedure for measuring the weight of the fetus and this has the high issue rate as there is lot of manual process involved in it. A baby in the womb is measured from head to bottom and from 12 weeks to till the delivery time. Throughout 20 weeks of empirical assessment, we will be getting a precise estimation of how the foetus is shaping up in the future. Traditional studies (Karen, 2016) like the below table 1.1 exist for physicians for reference purpose. However, accurate measures can help in better predictions and this leads to even more accurate treatment.