Prediction of Larger-than-gestational Age Based on Weak Supervised Learning

Zhi-rui WANG, Hong-wei LI

Abstract


Larger-than-gestational age prone to a variety of complications, and easy to produce all kinds of maternal injuries. It is of great significance to establish a prediction model for larger-than-gestational age for early diagnosis and intervention. In this study, data records of newborn fetuses collected between 2010 and 2013 were used as samples which included some unlabeled and inaccurately labeled records. Weak supervised learning technology in machine learning was used to predict diseases of larger-than-gestational age. The recall rate of the final prediction model was 0.82, the accuracy rate was 0.965, and the area under the curve was 0.89. Experimental results show that the proposed method is effective. In addition, this study verified that indicators such as parental body mass index, smoking status, creatinine, and hemoglobin were associated with larger-than-gestational age.

Keywords


Disease prediction, Machine learning, Weak supervised learning


DOI
10.12783/dtcse/cscbd2019/30082

Full Text:

PDF

Refbacks

  • There are currently no refbacks.