Paper Info
Title
CNN-based Estimation of Abdominal Circumference from Ultrasound images.
Abstract
The obstetrics and gynecology ultrasound diagnosis is routinely used to check fetal biometry, and due to its time-consuming routine process, there has been great demand of automatic estimation. Automated analysis of ultrasound images is complicated because ultrasound images are patient-specific, operator-dependent, and machine specific. Among fetal biometry, abdominal circumference (AC) is more difficult to make accurate measurement automatically because abdomen has low contrast against surroundings, non-uniform contrast and irregular shape compared to other parameters. This paper proposes a framework for estimation of the fetal AC from 2D ultrasound data by a specially designed convolutional neural network (CNN) which takes account of doctorsu0027 decision process, anatomical structure, and the characteristics of ultrasound image. The proposed framework uses CNN to classify ultrasound images (stomach bubble, amniotic fluid, and umbilical vein) and the Hough transform for the measurement of the AC. We tested the proposed method using clinical ultrasound data acquired from 10 pregnant women. Experimental results showed that, with relatively small training samples, the proposed CNN provided sufficient classification results for AC estimation through the Hough transform. This framework showed good performance on most cases and even for ultrasound images deteriorated by shadowing artifacts. However, for oversized fetus cases, when amniotic fluid is not seen or abdominal area was distorted, it could not estimate correct AC.
Year
Venue
Field
2017
arXiv: Computer Vision and Pattern Recognition
Nuclear medicine,Abdominal circumference,Computer science,Ultrasound
DocType
Volume
Citations 
Journal
abs/1702.02741
0
PageRank 
References 
Authors
0.34
12
6
Name
Order
Citations
PageRank
Jaeseong Jang100.68
Ja-Young Kwon211.05
Bukweon Kim311.39
Sung Min Lee400.68
Yejin Park500.34
Jin Keun Seo637658.65