Human-level Performance On Automatic Head Biometrics In Fetal Ultrasound Using Fully Convolutional Neural Networks

  title={Human-level Performance On Automatic Head Biometrics In Fetal Ultrasound Using Fully Convolutional Neural Networks},
  author={Matthew Sinclair and Christian F. Baumgartner and Jacqueline Matthew and Wenjia Bai and Juan Cerrolaza Martinez and Yuanwei Li and Sandra Smith and Caroline L. Knight and Bernhard Kainz and Joseph V. Hajnal and Andrew P. King and Daniel Rueckert},
  journal={2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)},
Measurement of head biometrics from fetal ultrasonography images is of key importance in monitoring the healthy development of fetuses. However, the accurate measurement of relevant anatomical structures is subject to large inter-observer variability in the clinic. To address this issue, an automated method utilizing Fully Convolutional Networks (FCN) is proposed to determine measurements of fetal head circumference (HC) and biparietal diameter (BPD). An FCN was trained on approximately 2000 2D… 

Figures and Tables from this paper

Automatic Measurement of Fetal Head Biometry from Ultrasound Images Using Deep Neural Networks

A deep learning-based method for automatic measurement of the BPD and HC based on the segmentation of the fetal head from ultrasound images based on an efficient convolutional network architecture, named multi-feature pyramid Unet (MFP-Unet).

Deep learning fetal ultrasound video model match human observers in biometric measurements

It is argued that FUVAI has the potential to assist sonographers who perform fetal biometric measurements in clinical settings by providing them with suggestions regarding the best measuring frames, along with automated measurements.

Improving Fetal Head Contour Detection by Object Localisation with Deep Learning

A deep learning-based method to segment fetal head from ultrasound images by incorporating an object localisation in a framework developed for segmentation purpose to improve the segmentation accuracy achieved by fully convolutional network.

Cascaded Transforming Multi-task Networks For Abdominal Biometric Estimation from Ultrasound

The proposed method to annotate the abdomen in 2D US images and measure AC using a shape-aware, multi-task deep convolutional neural network in a cascaded model framework significantly improves on the baseline model, while running at 8fps which could aid clinical workflow.

Confident Head Circumference Measurement from Ultrasound with Real-time Feedback for Sonographers

A novel probabilistic Deep Learning approach for real-time automated estimation of fetal HC measurements that feeds back statistics on measurement robustness to inform users how confident a deep neural network is in evaluating suitable views acquired during free-hand ultrasound examination.

Convolutional Neural Network to Detect and Measure Fetal Skull Circumference in Ultrasound Imaging

A new method for the segmentation of two-dimensional ultrasound images of fetal skulls based on a V-Net architecture called Fully Convolutional Neural Network - Combination (VNet-c) is presented, overcoming states-of-the-art methods.

FetalNet: Multi-task deep learning framework for fetal ultrasound biometric measurements

An end-to-end multi-task neural network called FetalNet with an attention mechanism and stacked module for spatio-temporal fetal ultrasound scan video analysis to simultaneously localize, classify and measure the fetal body parts is proposed.

Ensemble Transfer Learning for Fetal Head Analysis: From Segmentation to Gestational Age and Weight Prediction

This work presents a full end-to-end framework for segmenting, measuring, and estimating fetal gestational age and weight based on two-dimensional ultrasound images of the fetal head based on an optimized ensemble transfer learning model.

RDHCformer: Fusing ResDCN and Transformers for Fetal Head Circumference Automatic Measurement in 2D Ultrasound Images

A new one-stage network for rotating elliptic object detection based on anchor-free method, which is also an end-to-end network for fetal HC auto-measurement that no need for any post-processing is proposed.

Dilated Squeeze-and-Excitation U-Net for Fetal Ultrasound Image Segmentation

  • D. QiaoF. Zulkernine
  • Medicine
    2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB)
  • 2020
Deep learning encode-decode architecture is used to segment the fetal skull boundary and fetal skull for fetal HC measurement and a network based on U-Net is modified due to its outstanding performance in biomedical image analysis is modified.



Automatic Fetal Head Circumference Measurement in Ultrasound Using Random Forest and Fast Ellipse Fitting

A learning-based framework that used prior knowledge and employed a fast ellipse fitting method (ElliFit) to measure HC automatically and demonstrated that the method shows great promise for applications in clinical practice.

Detection and Measurement of Fetal Anatomies from Ultrasound Images using a Constrained Probabilistic Boosting Tree

A novel system for fast automatic detection and measurement of fetal anatomies that directly exploits a large database of expert annotated fetal anatomical structures in ultrasound images that learns automatically to distinguish between the appearance of the object of interest and background by training a constrained probabilistic boosting tree classifier.

SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand Ultrasound

A novel method based on convolutional neural networks is proposed, which can automatically detect 13 fetal standard views in freehand 2-D ultrasound data as well as provide a localization of the fetal structures via a bounding box while providing optimal output for the localization task.

Cascaded Fully Convolutional Networks for automatic prenatal ultrasound image segmentation

A cascaded framework for fully automatic US image segmentation that can bridge severe boundary incompleteness and achieves the best segmentation accuracy when compared with state-of-the-art methods is proposed.

Fully convolutional networks for semantic segmentation

The key insight is to build “fully convolutional” networks that take input of arbitrary size and produce correspondingly-sized output with efficient inference and learning.

Intra‐ and interobserver variability in fetal ultrasound measurements

  • I. SarrisC. Ioannou A. Papageorghiou
  • Medicine
    Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology
  • 2012
To assess intra‐ and interobserver variability of fetal biometry measurements throughout pregnancy, a large number of patients with high‐risk pregnancies are randomly assigned to a baseline or sham pregnancy.

Direct least squares fitting of ellipses

This paper presents a new efficient method for fitting ellipses to scattered data that is ellipse-specific so that even bad data will always return an ellipso, and can be solved naturally by a generalized eigensystem.

Direct Least Square Fitting of Ellipses

This work presents a new efficient method for fitting ellipses to scattered data by minimizing the algebraic distance subject to the constraint 4ac-b/sup 2/=1, which incorporates the ellipticity constraint into the normalization factor.

Inequalities for the Perimeter of an Ellipse

Let a, b denote the lengths of the semi-major and -minor axes of an ellipse with eccentricity e = (1/a) √ a2 − b2 and whose perimeter is L(a, b) . The problem of approximating L(a, b) is an ancient

Evaluation and Comparison of Current Fetal Ultrasound Image Segmentation Methods for Biometric Measurements: A Grand Challenge

Evaluation results of the methods submitted to Challenge US: Biometric Measurements from Fetal Ultrasound Images, a segmentation challenge held at the IEEE International Symposium on Biomedical Imaging 2012, show the femur sub-challenge had inferior performance to the head sub-Challenge, and several groups produced results that could be potentially used in clinical settings, with comparable performance to manual delineations.