arXiv subjects

Jian Zhuang

Publications and source records attributed to Jian Zhuang.

24 records · Page 2Linked to original sources

arXiv2019

MSU-Net: Multiscale Statistical U-Net for Real-time 3D Cardiac MRI Video Segmentation

Cardiac magnetic resonance imaging (MRI) is an essential tool for MRI-guided surgery and real-time intervention. The MRI videos are expected to be segmented on-the-fly in real practice. However, existing segmentation methods would suffer from drastic accuracy loss when modified for speedup. In this work, we propose Multiscale Statistical U-Net (MSU-Net) for real-time 3D MRI video segmentation in cardiac surgical guidance. Our idea is to model the input samples as multiscale canonical form distributions for speedup, while the spatio-temporal correlation is still fully utilized. A parallel statistical U-Net is then designed to efficiently process these distributions. The fast data sampling and efficient parallel structure of MSU-Net endorse the fast and accurate inference. Compared with vanilla U-Net and a modified state-of-the-art method GridNet, our method achieves up to 268% and 237% speedup with 1.6% and 3.6% increased Dice scores.

eess.IV↗

arXiv2019

Accurate Congenital Heart Disease Model Generation for 3D Printing

3D printing has been widely adopted for clinical decision making and interventional planning of Congenital heart disease (CHD), while whole heart and great vessel segmentation is the most significant but time-consuming step in the model generation for 3D printing. While various automatic whole heart and great vessel segmentation frameworks have been developed in the literature, they are ineffective when applied to medical images in CHD, which have significant variations in heart structure and great vessel connections. To address the challenge, we leverage the power of deep learning in processing regular structures and that of graph algorithms in dealing with large variations and propose a framework that combines both for whole heart and great vessel segmentation in CHD. Particularly, we first use deep learning to segment the four chambers and myocardium followed by the blood pool, where variations are usually small. We then extract the connection information and apply graph matching to determine the categories of all the vessels. Experimental results using 683D CT images covering 14 types of CHD show that our method can increase Dice score by 11.9% on average compared with the state-of-the-art whole heart and great vessel segmentation method in normal anatomy. The segmentation results are also printed out using 3D printers for validation.

eess.IV↗

arXiv2019

Machine Vision Guided 3D Medical Image Compression for Efficient Transmission and Accurate Segmentation in the Clouds

Cloud based medical image analysis has become popular recently due to the high computation complexities of various deep neural network (DNN) based frameworks and the increasingly large volume of medical images that need to be processed. It has been demonstrated that for medical images the transmission from local to clouds is much more expensive than the computation in the clouds itself. Towards this, 3D image compression techniques have been widely applied to reduce the data traffic. However, most of the existing image compression techniques are developed around human vision, i.e., they are designed to minimize distortions that can be perceived by human eyes. In this paper we will use deep learning based medical image segmentation as a vehicle and demonstrate that interestingly, machine and human view the compression quality differently. Medical images compressed with good quality w.r.t. human vision may result in inferior segmentation accuracy. We then design a machine vision oriented 3D image compression framework tailored for segmentation using DNNs. Our method automatically extracts and retains image features that are most important to the segmentation. Comprehensive experiments on widely adopted segmentation frameworks with HVSMR 2016 challenge dataset show that our method can achieve significantly higher segmentation accuracy at the same compression rate, or much better compression rate under the same segmentation accuracy, when compared with the existing JPEG 2000 method. To the best of the authors' knowledge, this is the first machine vision guided medical image compression framework for segmentation in the clouds.

cs.CV↗

arXiv2018

Design of TDC ASIC based on Temperature Compensation

.On the basis of requirement of CSNS, we designed a TDC chip with temperature compensation function in this paper, which employed TSMC 180nm process. Using delay unit bufx8 as the major method, delay lines in each level delayed input signal line through the bufx8 unit to realize fundamental measurement function. The time intervals of two fixed delay standard pulses did not change with temperature variation via intra-chip phase-locked loop. After that, the two standard pulses were sent to TDC internal delay line and measured their values. Then the measured values and standard values were compared. According to the result of comparing and decision switch, the structure of delay lines was reconstructed and their levels were recorded at the same time. We could ensure that the total length of the effective delay line were close to clock cycle as much as possible under the current temperature. The chip was tested after the completion of design. It was found that the time resolution of TDC ASIC was 73ps under 1.8V power supply at room temperature while the time resolutions were 103ps and 62ps at 85$^\circ$ and 0$^\circ$, respectively.

physics.ins-det↗

arXiv2018

Data Acquisition System for CSNS Neutron Beam Monitor

In Chinese Spallation Neutron Source (CSNS), proton beam is used to hit metal tungsten target, and then high flux neutrons are generated for experiments on instruments. For neutron flux spectrum correction, boron-coated GEM, lithium glass and 3He are used as neutron beam monitor in instruments. To be integrated into neutron instrument control, a new DAQ software for neutron beam monitor is developed, called NEROS (Neutron Event Readout System). NEROS is based on EPICS V4 and a unified data format for CSNS neutron beam monitor is defined. The framework and software design of NEROS is introduced in this paper, including real-time data readout, data processing, data visualization and data storage in Nexus format. The performance is evaluated through offline test, X-ray test and 252Cf test. The deployment of NEROS and its running result in CSNS instrument commissioning are also introduced in this paper.

physics.ins-det↗

arXiv2018

The Proton Beam Real-time Monitor System in CSNS

In Chinese Spallation Neutron Source (CSNS), proton beam is used to hit metal tungsten target, and then high flux neutron are generated for experiments on instruments. For neutron flux spectrum correction, the current of proton beam is for each instrument. A real time monitor system is developed in CSNS, to monitor, broadcast and record each pulse of proton. Each proton pulse charge is measured and marked with high-precision timestamp. Then, the result of measurement will be broadcasted to control room and each neutron instrument. In control room, the proton charge of each pulse is listened and stored in database by agent program for offline use. The high-precision timestamp can be used to proton charge and neutron data alignment in time scale. The architecture of proton beam monitor system is introduced in this paper. And the performance of this system is evaluated in this paper.

physics.ins-det↗