loading page

A Deep Convolutional Neural Network based Hybrid Framework for Fetal Head Standard Plane Identification
  • +9
  • Jingyu Ye,
  • Ruizhi Liu,
  • Bin Zou,
  • Hongyang Zhang,
  • nianji zhan,
  • Cong Han,
  • Ying Yang,
  • Hongguo Zhang,
  • Jian Guo,
  • Fang Chen,
  • Shida Zhu,
  • Shucheng Hua
Jingyu Ye
BGI-Shenzhen

Corresponding Author:newzolo@gmail.com

Author Profile
Ruizhi Liu
Jilin University First Hospital
Author Profile
Bin Zou
CNGB
Author Profile
Hongyang Zhang
Jilin University First Hospital
Author Profile
nianji zhan
Author Profile
Cong Han
Jilin University First Hospital
Author Profile
Ying Yang
BGI-Shenzhen
Author Profile
Hongguo Zhang
Jilin University First Hospital
Author Profile
Jian Guo
BGI-Shenzhen
Author Profile
Fang Chen
BGI-Shenzhen
Author Profile
Shida Zhu
BGI-Shenzhen
Author Profile
Shucheng Hua
Jilin University First Hospital Department of Respiratory Medicine
Author Profile

Abstract

As considered to be less risky, less expensive, and more convenient than radiological examinations, ultrasound has been routinely employed in prenatal exams for the past decades. However, the quality of acquired ultrasound samples, i.e., ultrasound images or videos, and the further diagnosis is crucially depended on the sonographer. At the meantime, there are an extremely limited number of experienced sonographer available for the fetal ultrasound screening. Therefore, to reduce the workload of sonographers, and to promote the quality of fetal ultrasound screening, a deep convolutional neural network based framework is proposed for automatically differentiating five types of fetal head ultrasound standard planes, i.e., Transventricular plane (TV), Transthalamic plane (TT), Transcerebellar plane (TC), Coronal view of eyes (Eyes), Coronal view of nose (Nose), and other non-standard fetal head ultrasound images (Background). A dataset consists of 19928 fetal ultrasound images is applied for the model training and performance evaluation. By combining object detection network, object classification network, and model stacking technique, the proposed framework achieves the state-of-the-art performance with the average accuracy of 89.61% and the average F-1 score of 89.61%.