Secret History: Return of the Black Death Channel 4, 7-8pm In 1348 the Black Death swept through London, killing people within days of the appearance of their first symptoms. Exactly how many died, and why, has long been a mystery. This Secret History documentary follows experts as they pick through the evidence and reveal why the plague killed on such a scale. And they ask, what might be coming next?

Medscape

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

IEEE Transactions on Pattern Analysis and Machine Intelligence

X-ray computer tomography (CT) is fast becoming an accepted tool within the materials science community for the acquisition of 3D images. Here the authors review the current state of the art as CT transforms from a qualitative diagnostic tool to a quantitative one. Our review considers first the image acquisition process, including the use of iterative reconstruction strategies suited to specific segmentation tasks and emerging methods that provide more insight (e.g. fast and high resolution imaging, crystallite (grain) imaging) than conventional attenuation based tomography. Methods and shortcomings of CT are examined for the quantification of 3D volumetric data to extract key topological parameters such as phase fractions, phase contiguity, and damage levels as well as density variations. As a non-destructive technique, CT is an ideal means of following structural development over time via time lapse sequences of 3D images (sometimes called 3D movies or 4D imaging). This includes information nee...

https://www.tandfonline.com/doi/pdf/10.1179/1743280413Y.0000000023

Quantitative X-ray tomography

In order to reduce the potential radiation risk, low-dose CT has attracted an increasing attention. However, simply lowering the radiation dose will significantly degrade the image quality. In this paper, we propose a new noise reduction method for low-dose CT via deep learning without accessing original projection data. A deep convolutional neural network is here used to map low-dose CT images towards its corresponding normal-dose counterparts in a patch-by-patch fashion. Qualitative results demonstrate a great potential of the proposed method on artifact reduction and structure preservation. In terms of the quantitative metrics, the proposed method has showed a substantial improvement on PSNR, RMSE and SSIM than the competing state-of-art methods. Furthermore, the speed of our method is one order of magnitude faster than the iterative reconstruction and patch-based image denoising methods.

Low-dose CT via convolutional neural network

The invention discloses a segmentation method of dissected aorta in CT images based on a convolution neural network. The invention provides a CT dissecting aorta segmentation algorithm combining a three-dimensional convolution neural network and a two-dimensional convolution neural network, wherein the three-dimensional volume data is divided into two parts by using the three-dimensional convolution neural network, and the two parts are segmented by using two two-dimensional convolution neural networks respectively to obtain a final segmentation result. The invention can effectively segment the dissected aorta from a CT image containing the dissected aorta. It overcomes the shortcomings of poor segmentation precision caused by the contradiction between the image resolution and the GPU memory capacity because only a three-dimensional full convelution neural network is used, and the instability of segmentation caused by loss of three-dimensional information when only the two-dimensionalconvolutional neural network is used, so it has good segmentation effect.

A segmentation method of dissected aorta in CT images based on a convolution neural network

Dynamic computed tomography (CT) imaging aims to access the kinetics of the moving organs. In cardiac imaging, the interest lies in the possibility of obtaining anatomic and functional information on the heart and the coronaries during the same examination. However, segmentation, reconstruction and registration algorithms need to be developed for diagnostic purposes. We propose thus to built a 3D heart model from multi-slice spiral computed tomography (MSCT) dynamic sequences to facilitate the evaluation of these algorithms. The model building relies on semi-automatic segmentation techniques based on deformable models such as fast marching and active contours. Shape-based interpolation and marching cube algorithms are then used for the 3D surface reconstruction

A 3D Static Heart Model From a MSCT Data Set

Coronary artery diseases are the most common type of heart disease. Early detection and quantification of coronary plaques is therefore of high interest. CTA has rapidly emerged, and is nowadays widely used in clinical practice. A calcification detection and quantification method is proposed, which can detect the calcium plaque and quantify the stenosis of coronary artery in CT images. Firstly, the responses of the DoG operator in the center on each slice of the artery are obtained to estimate the diameter of the vessel. Then, the slice is cropped according to the feature size. Secondly, the Fuzzy C-means is applied to classify the cropped image. Finally, the calcium plaques are extracted with their intensity and the stenosis of the artery caused by the calcified plaque is quantified according to the estimated cross-section area of the vessel. Experiments show that the proposed method can detect and quantify the calcium plaques accurately.

Calcified coronary plaques detection in CTA based-on automatic scale selection and fuzzy C means

As a particular case of the finite mixture model, Rayleigh mixture model (RMM) is considered as a useful tool for medical ultrasound (US) image segmentation. However, conventional RMM relies on intensity distribution only and does not take any spatial information into account that leads to misclassification on boundaries and inhomogeneous regions. The authors proposed an improved RMM with neighbour (RMMN) information to solve this problem by introducing neighbourhood information through a mean template. The incorporation of the spatial information made RMMN more robust to noise on the boundaries. The size of the window which incorporates neighbour information was resized adaptively according to the local gradient distribution. They evaluated their model on experiments on synthetic data and real US images used by high-intensity focused ultrasound therapy. On this data, they demonstrated that the proposed model outperforms several state-of-the-art methods in terms of both segmentation accuracy and computation time.

/pdf/fast-segmentation-of-ultrasound-images-by-incorporating-akqdy0zcdg.pdf

Fast segmentation of ultrasound images by incorporating spatial information into Rayleigh mixture model

Federated learning (FL) has shown value in multi-center multi-sequence cardiac magnetic resonance (CMR) segmentation, due to imbalanced CMR distributions and privacy preservation in clinical practice. However, the larger heterogeneity among multi-center multi-sequence CMR brings challenges to the FL framework: (1) Representation bias in the model fusion. The FL server model, which is generated by an average fusion of heterogeneous client models, is biased to representation close to the mean distribution and away from the long-distance distribution. Hence, the FL has poor representation ability. (2) Optimization stop in the model replacing. The heterogeneous server model replacing client model in FL directly causes the long-distance clients to utilize worse optimization to replace the original optimization. The client has to recover the optimization with the worse initialization, hence it lacks the continuous optimization ability. In this work, a cross-center cross-sequence medical image segmentation FL framework (FedCRLD) is proposed for the first time to facilitate multi-center multi-sequence CMR segmentation. (1) The contrastive re-localization module (CRL) of FedCRLD enables the correct representation from the heterogeneous model by embedding a novel contrastive difference metric of mutual information into a cross-attention localization transformer to transfer client-correlated knowledge from server model without bias. (2) The momentum distillation strategy (MD) of FedCRLD enables continuous optimization by conducting self-training on a dynamically updated client momentum bank to refine optimization by local correct optimization history. FedCRLD is validated on 420 CMR images 6 clients from 2 public datasets scanned by different hospitals, devices and contrast agents. Our FedCRLD achieves superior performance on multi-center multi-sequence CMR segmentation (average dice 85.96%). https://github.com/JerryQseu/FedCRLD. 

Guanyu Yang

Papers

A segmentation method of dissected aorta in CT images based on a convolution neural network

A 3D Static Heart Model From a MSCT Data Set

Calcified coronary plaques detection in CTA based-on automatic scale selection and fuzzy C means

Fast segmentation of ultrasound images by incorporating spatial information into Rayleigh mixture model

Contrastive Re-localization and History Distillation in Federated CMR Segmentation