Towards Whole Placenta Segmentation At Late Gestation Using Multi-View Ultrasound Images
read more
Citations
A Review on Deep-Learning Algorithms for Fetal Ultrasound-Image Analysis
Ultrasound Medical Imaging Techniques: A Survey
Towards Standardized Acquisition With a Dual-Probe Ultrasound Robot for Fetal Imaging
A Multi-task Approach Using Positional Information for Ultrasound Placenta Segmentation.
Deep Learning strategies for Ultrasound in Pregnancy.
References
U-Net: Convolutional Networks for Biomedical Image Segmentation
U-Net: Convolutional Networks for Biomedical Image Segmentation
A survey on deep learning in medical image analysis
Towards Automated Semantic Segmentation in Prenatal Volumetric Ultrasound
Fully automated, real-time 3D ultrasound segmentation to estimate first trimester placental volume using deep learning
Related Papers (5)
Weakly supervised learning of placental ultrasound images with residual networks
Frequently Asked Questions (17)
Q2. What are the future works in "Towards whole placenta segmentation at late gestation using multi-view ultrasound images" ?
Future work will include the refinement of the segmentation method. Their method enables future work for the comparison of US and MR placenta in more detail.
Q3. How do the authors acquire multiple US images?
The authors acquire multiple US images using an in-house US signal multiplexer which allows to connect multiple Philips X6-1 probes to a Philips EPIQ V7 US system.
Q4. What is the primary imaging modality used to monitor fetal development?
State-of-the-art segmentation methods using convolutional neural networks (CNNs) have been used in [7, 4] and in [13] additionally for the fetus and the gestational sac.
Q5. How is the segmentation of an object learned?
In a supervised CNN approach, the segmentation of an object is learned only driven by the data from a training dataset T = {(In, Sn), n = 1, . . . , N}, with images
Q6. How many images were used to test the pipeline?
The authors used a dataset of 127 3D US images to test their pipeline, which were selected from 4D (3D+time) image streams from 30 patients covering different parts of the placenta.
Q7. What is the effect of the US signal multiplexer?
In effect, image points with a strong signal (to correct for shadow artifacts) and at a position close to the center of the US frustum (where the quality of the image is typically the best) will receive higher weights.
Q8. What is the way to visualize the whole placenta in one volume?
Multiple probes, as described above, or placenta sweeps using an appropriate registration or tracking method to align the images, can be used to visualize the whole placenta in one US volume.
Q9. What is the resulting segmentation of the whole placenta?
The resulting segmentations for the individual images are aligned and fused using maximum intensity voting to obtain the segmentation of the whole placenta.
Q10. What is the advantage of multi-view imaging?
The resulting image has an extended FoV, and viewdependent artifacts such as shadows can be minimized through the additional signal information from multiple views [14].
Q11. What is the main reason why ultrasound is the standard screening tool?
US is still the standard screening tool because, in contrast to MRI, it is performed in real-time, is relatively inexpensive, and widely available.
Q12. How many probes can be used to acquire multiple images?
The multiplexer switches rapidly between up to three probes so that images from each probe are acquired in a time-interleaved fashion.
Q13. What is the intensity of a point x of the fused image?
In other words, the intensity of a point x of the fused image is calculated by the weighted mean of corresponding points in the single images.
Q14. What is the primary imaging modality to study the placenta in utero?
While the fetal body, especially the fetal brain, are subjects of intensive research, only few methods exist to study the placenta in utero [11].
Q15. What is the funding for this work?
This work was supported by the Wellcome Trust IEH Award [102431], by the Wellcome/EPSRC Centre for Medical Engineering [WT203148/Z/16/Z] and by the National Institute for Health Research (NIHR)Biomedical Research Centre at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London.
Q16. What is the primary imaging modality for assessing placental development?
These methods focused on early pregnancies between 10-14 weeks of gestational age (GA), when the placenta is small enough to fit in the limited FoV of US.
Q17. How can the authors extract the whole placenta from multi-view images?
The authors present a voxel-wise image fusion method to combine the images and to reduce view-dependent artifacts, and compare four approaches based on CNNs to extract the whole placenta from the multi-view images.