Automatic Segmentation and Location Learning of Neonatal Cerebral Ventricles in 3D Ultrasound Data Combining CNN and CPPN

by   Matthieu Martin, et al.

Preterm neonates are highly likely to suffer from ventriculomegaly, a dilation of the Cerebral Ventricular System (CVS). This condition can develop into life-threatening hydrocephalus and is correlated with future neuro-developmental impairments. Consequently, it must be detected and monitored by physicians. In clinical routing, manual 2D measurements are performed on 2D ultrasound (US) images to estimate the CVS volume but this practice is imprecise due to the unavailability of 3D information. A way to tackle this problem would be to develop automatic CVS segmentation algorithms for 3D US data. In this paper, we investigate the potential of 2D and 3D Convolutional Neural Networks (CNN) to solve this complex task and propose to use Compositional Pattern Producing Network (CPPN) to enable the CNNs to learn CVS location. Our database was composed of 25 3D US volumes collected on 21 preterm nenonates at the age of 35.8 ± 1.6 gestational weeks. We found that the CPPN enables to encode CVS location, which increases the accuracy of the CNNs when they have few layers. Accuracy of the 2D and 3D CNNs reached intraobserver variability (IOV) in the case of dilated ventricles with Dice of 0.893 ± 0.008 and 0.886 ± 0.004 respectively (IOV = 0.898 ± 0.008) and with volume errors of 0.45 ± 0.42 cm^3 and 0.36 ± 0.24 cm^3 respectively (IOV = 0.41 ± 0.05 cm^3). 3D CNNs were more accurate than 2D CNNs in the case of normal ventricles with Dice of 0.797 ± 0.041 against 0.776 ± 0.038 (IOV = 0.816 ± 0.009) and volume errors of 0.35 ± 0.29 cm^3 against 0.35 ± 0.24 cm^3 (IOV = 0.2 ± 0.11 cm^3). The best segmentation time of volumes of size 320 × 320 × 320 was obtained by a 2D CNN in 3.5 ± 0.2 s.


page 3

page 4

page 5

page 8

page 11

page 12


Tracked 3D Ultrasound and Deep Neural Network-based Thyroid Segmentation reduce Interobserver Variability in Thyroid Volumetry

Background: Thyroid volumetry is crucial in diagnosis, treatment and mon...

SUMNet: Fully Convolutional Model for Fast Segmentation of Anatomical Structures in Ultrasound Volumes

Ultrasound imaging is generally employed for real-time investigation of ...

Fetal Ultrasound Image Segmentation for Measuring Biometric Parameters Using Multi-Task Deep Learning

Ultrasound imaging is a standard examination during pregnancy that can b...

Dilatation of Lateral Ventricles with Brain Volumes in Infants with 3D Transfontanelle US

Ultrasound (US) can be used to assess brain development in newborns, as ...

Large-Scale Analysis of Iliopsoas Muscle Volumes in the UK Biobank

Psoas muscle measurements are frequently used as markers of sarcopenia a...

A regression framework to head-circumference delineation from US fetal images

Background and Objectives Measuring head-circumference (HC) length from ...

A two-stage 3D Unet framework for multi-class segmentation on full resolution image

Deep convolutional neural networks (CNNs) have been intensively used for...