Searcharxiv⌕ Search

arXiv subjects

Boyu Zhang

Publications and source records attributed to Boyu Zhang.

87 records · Page 5Linked to original sources
arXiv2019

Fuzzy Semantic Segmentation of Breast Ultrasound Image with Breast Anatomy Constraints

Breast cancer is one of the most serious disease affecting women's health. Due to low cost, portable, no radiation, and high efficiency, breast ultrasound (BUS) imaging is the most popular approach for diagnosing early breast cancer. However, ultrasound images are low resolution and poor quality. Thus, developing accurate detection system is a challenging task. In this paper, we propose a fully automatic segmentation algorithm consisting of two parts: fuzzy fully convolutional network and accurately fine-tuning post-processing based on breast anatomy constraints. In the first part, the image is preprocessed by contrast enhancement, and wavelet features are employed for image augmentation. A fuzzy membership function transforms the augmented BUS images into fuzzy domain. The features from convolutional layers are processed using fuzzy logic as well. The conditional random fields (CRFs) post-process the segmentation result. The location relation among the breast anatomy layers is utilized to improve the performance. The proposed method is applied to the dataset with 325 BUS images, and achieves state-of-art performance compared with that of existing methods with true positive rate 90.33%, false positive rate 9.00%, and intersection over union (IoU) 81.29% on tumor category, and overall intersection over union (mIoU) 80.47% over five categories: fat layer, mammary layer, muscle layer, background, and tumor.

cs.CV↗
arXiv2019

Breast Anatomy Enriched Tumor Saliency Estimation

Breast cancer investigation is of great significance, and developing tumor detection methodologies is a critical need. However, it is a challenging task for breast ultrasound due to the complicated breast structure and poor quality of the images. In this paper, we propose a novel tumor saliency estimation model guided by enriched breast anatomy knowledge to localize the tumor. Firstly, the breast anatomy layers are generated by a deep neural network. Then we refine the layers by integrating a non-semantic breast anatomy model to solve the problems of incomplete mammary layers. Meanwhile, a new background map generation method weighted by the semantic probability and spatial distance is proposed to improve the performance. The experiment demonstrates that the proposed method with the new background map outperforms four state-of-the-art TSE models with increasing 10% of F_meansure on the BUS public dataset.

cs.CV↗
arXiv2019

Efficient Inference of CNNs via Channel Pruning

The deployment of Convolutional Neural Networks (CNNs) on resource constrained platforms such as mobile devices and embedded systems has been greatly hindered by their high implementation cost, and thus motivated a lot research interest in compressing and accelerating trained CNN models. Among various techniques proposed in literature, structured pruning, especially channel pruning, has gain a lot focus due to 1) its superior performance in memory, computation, and energy reduction; and 2) it is friendly to existing hardware and software libraries. In this paper, we investigate the intermediate results of convolutional layers and present a novel pivoted QR factorization based channel pruning technique that can prune any specified number of input channels of any layer. We also explore more pruning opportunities in ResNet-like architectures by applying two tweaks to our technique. Experiment results on VGG-16 and ResNet-50 models with ImageNet ILSVRC 2012 dataset are very impressive with 4.29X and 2.84X computation reduction while only sacrificing about 1.40\% top-5 accuracy. Compared to many prior works, the pruned models produced by our technique require up to 47.7\% less computation while still achieve higher accuracies.

cs.CV↗
arXiv2019

Instanton Floer homology for sutured manifolds with tangles

We prove an excision theorem for the singular instanton Floer homology that allows the excision surfaces to intersect the singular locus. This is an extension of the non-singular excision theorem by Kronheimer and Mrowka and the genus-zero singular excision theorem by Street. We use the singular excision theorem to define an instanton Floer homology theory for sutured manifolds with tangles. As applications, we prove that the annular Khovanov homology (1) detects the unlink, (2) detects the closure of the trivial braid, and (3) distinguishes braid closures from other links; we also prove that the annular instanton Floer homology detects the Thurston norm of meridional surfaces.

math.GT↗
arXiv2019

Tumor Saliency Estimation for Breast Ultrasound Images via Breast Anatomy Modeling

Tumor saliency estimation aims to localize tumors by modeling the visual stimuli in medical images. However, it is a challenging task for breast ultrasound due to the complicated anatomic structure of the breast and poor image quality; and existing saliency estimation approaches only model generic visual stimuli, e.g., local and global contrast, location, and feature correlation, and achieve poor performance for tumor saliency estimation. In this paper, we propose a novel optimization model to estimate tumor saliency by utilizing breast anatomy. First, we model breast anatomy and decompose breast ultrasound image into layers using Neutro-Connectedness; then utilize the layers to generate the foreground and background maps; and finally propose a novel objective function to estimate the tumor saliency by integrating the foreground map, background map, adaptive center bias, and region-based correlation cues. The extensive experiments demonstrate that the proposed approach obtains more accurate foreground and background maps with the assistance of the breast anatomy; especially, for the images having large or small tumors; meanwhile, the new objective function can handle the images without tumors. The newly proposed method achieves state-of-the-art performance when compared to eight tumor saliency estimation approaches using two breast ultrasound datasets.

cs.CV↗
arXiv2019

Energy-efficient domain wall motion governed by the interplay of helicity-dependent optical effect and spin-orbit torque

Spin-orbit torque provides a powerful means of manipulating domain walls along magnetic wires. However, the current density required for domain wall motion is still too high to realize low power devices. Here we experimentally demonstrate helicity-dependent domain wall motion by combining synchronized femtosecond laser pulses and short current pulses in Co/Ni/Co ultra-thin film wires with perpendicular magnetization. Domain wall can remain pinned under one laser circular helicity while depinned by the opposite circular helicity. Thanks to the all-optical helicity-dependent effect, the threshold current density due to spin-orbit torque can be reduced by more than 50%. Based on this joint effect combining spin-orbit torque and helicity-dependent laser pulses, an optoelectronic logic-in-memory device has been experimentally demonstrated. This work enables a new class of low power spintronic-photonic devices beyond the conventional approach of all-optical switching or all-current switching for data storage.

cond-mat.mes-hall↗
arXiv2019

Domain-wall motion induced by spin transfer torque delivered by helicity-dependent femtosecond laser

In magnetic wires with perpendicular anisotropy, moving domain with only current or only circularly polarized light requires a high power. Here, we propose to reduce it by using both short current pulses and femtosecond laser pulses simultaneously. The wires were made out of perpendicularly magnetized film of Pt/Co/Ni/Co/Pt. The displacement of the domain wall is found to be dependent on the laser helicity. Based on a quantitative analysis of the current-induced domain wall motion, the spin orbit torque contribution can be neglected when compared to the spin transfer torque contribution. The effective field of the spin transfer torque is extracted from the pulsed field domain wall measurements. Finally, our result can be described using the Fatuzzo-Labrune model and considering the effective field due to the polarized laser beam, the effective field due to spin transfer torque, and the Gaussian temperature distribution of the laser spot.

cond-mat.mes-hall↗
arXiv2018

A Mixture of Expert Approach for Low-Cost Customization of Deep Neural Networks

The ability to customize a trained Deep Neural Network (DNN) locally using user-specific data may greatly enhance user experiences, reduce development costs, and protect user's privacy. In this work, we propose to incorporate a novel Mixture of Experts (MOE) approach to accomplish this goal. This architecture comprises of a Global Expert (GE), a Local Expert (LE) and a Gating Network (GN). The GE is a trained DNN developed on a large training dataset representative of many potential users. After deployment on an embedded edge device, GE will be subject to customized, user-specific data (e.g., accent in speech) and its performance may suffer. This problem may be alleviated by training a local DNN (the local expert, LE) on a small size customized training data to correct the errors made by GE. A gating network then will be trained to determine whether an incoming data should be handled by GE or LE. Since the customized dataset is in general very small, the cost of training LE and GN would be much lower than that of re-training of GE. The training of LE and GN thus can be performed at local device, properly protecting the privacy of customized training data. In this work, we developed a prototype MOE architecture for handwritten alphanumeric character recognition task. We use EMNIST as the generic dataset, LeNet5 as GE, and handwritings of 10 users as the customized dataset. We show that with the LE and GN, the classification accuracy is significantly enhanced over the customized dataset with almost no degradation of accuracy over the generic dataset. In terms of energy and network size, the overhead of LE and GN is around 2.5% compared to those of GE.

cs.CV↗
arXiv2018

A Hybrid Framework for Tumor Saliency Estimation

Automatic tumor segmentation of breast ultrasound (BUS) image is quite challenging due to the complicated anatomic structure of breast and poor image quality. Most tumor segmentation approaches achieve good performance on BUS images collected in controlled settings; however, the performance degrades greatly with BUS images from different sources. Tumor saliency estimation (TSE) has attracted increasing attention to solving the problem by modeling radiologists' attention mechanism. In this paper, we propose a novel hybrid framework for TSE, which integrates both high-level domain-knowledge and robust low-level saliency assumptions and can overcome drawbacks caused by direct mapping in traditional TSE approaches. The new framework integrated the Neutro-Connectedness (NC) map, the adaptive-center, the correlation and the layer structure-based weighted map. The experimental results demonstrate that the proposed approach outperforms state-of-the-art TSE methods.

cs.CV↗
arXiv2018

Computer-Aided Knee Joint Magnetic Resonance Image Segmentation - A Survey

Osteoarthritis (OA) is one of the major health issues among the elderly population. MRI is the most popular technology to observe and evaluate the progress of OA course. However, the extreme labor cost of MRI analysis makes the process inefficient and expensive. Also, due to human error and subjective nature, the inter- and intra-observer variability is rather high. Computer-aided knee MRI segmentation is currently an active research field because it can alleviate doctors and radiologists from the time consuming and tedious job, and improve the diagnosis performance which has immense potential for both clinic and scientific research. In the past decades, researchers have investigated automatic/semi-automatic knee MRI segmentation methods extensively. However, to the best of our knowledge, there is no comprehensive survey paper in this field yet. In this survey paper, we classify the existing methods by their principles and discuss the current research status and point out the future research trend in-depth.

cs.CV↗
arXiv2018

Automatic Breast Ultrasound Image Segmentation: A Survey

Breast cancer is one of the leading causes of cancer death among women worldwide. In clinical routine, automatic breast ultrasound (BUS) image segmentation is very challenging and essential for cancer diagnosis and treatment planning. Many BUS segmentation approaches have been studied in the last two decades, and have been proved to be effective on private datasets. Currently, the advancement of BUS image segmentation seems to meet its bottleneck. The improvement of the performance is increasingly challenging, and only few new approaches were published in the last several years. It is the time to look at the field by reviewing previous approaches comprehensively and to investigate the future directions. In this paper, we study the basic ideas, theories, pros and cons of the approaches, group them into categories, and extensively review each category in depth by discussing the principles, application issues, and advantages/disadvantages.

cs.CV↗
arXiv2017

Influence of heavy metal materials on magnetic properties of Pt/Co/heavy metal tri-layered structures

Pt/Co/heavy metal (HM) tri-layered structures with interfacial perpendicular magnetic anisotropy (PMA) are currently under intensive research for several emerging spintronic effects, such as spinorbit torque, domain wall motion, and room temperature skyrmions. HM materials are used as capping layers to generate the structural asymmetry and enhance the interfacial effects. For instance, the Pt/Co/Ta structure attracts a lot of attention as it may exhibit large Dzyaloshinskii-Moriya interaction. However, the dependence of magnetic properties on different capping materials has not been systematically investigated. In this paper, we experimentally show the interfacial PMA and damping constant for Pt/Co/HM tri-layered structures through time-resolved magneto-optical Kerr effect measurements as well as magnetometry measurements, where the capping HM materials are W, Ta, and Pd. We found that the Co/HM interface plays an important role on the magnetic properties. In particular, the magnetic multilayers with a W capping layer features the lowest effective damping value, which may be attributed to the different spin-orbit coupling and interfacial hybridization between Co and HM materials. Our findings allow a deep understanding of the Pt/Co/HM tri-layered structures. Such structures could lead to a better era of data storage and processing devices.

cond-mat.mes-hall↗
arXiv2017

Modulo 2 counting of Klein-bottle leaves in smooth taut foliations

This article proves that the parity of the number of Klein-bottle leaves in a smooth cooriented taut foliation is invariant under smooth deformations within taut foliations, provided that every Klein-bottle leaf involved in the counting has non-trivial linear holonomy.

math.GT↗
arXiv2015

In-Situ Data Analysis of Protein Folding Trajectories

The transition from petascale to exascale computers is characterized by substantial changes in the computer architectures and technologies. The research community relying on computational simulations is being forced to revisit the algorithms for data generation and analysis due to various concerns, such as higher degrees of concurrency, deeper memory hierarchies, substantial I/O and communication constraints. Simulations today typically save all data to analyze later. Simulations at the exascale will require us to analyze data as it is generated and save only what is really needed for analysis, which must be performed predominately in-situ, i.e., executed sufficiently fast locally, limiting memory and disk usage, and avoiding the need to move large data across nodes. In this paper, we present a distributed method that enables in-situ data analysis for large protein folding trajectory datasets. Traditional trajectory analysis methods currently follow a centralized approach that moves the trajectory datasets to a centralized node and processes the data only after simulations have been completed. Our method, on the other hand, captures conformational information in-situ using local data only while reducing the storage space needed for the part of the trajectory under consideration. This method processes the input trajectory data in one pass, breaks from the centralized approach of traditional analysis, avoids the movement of trajectory data, and still builds the global knowledge on the formation of individual $α$-helices or $β$-strands as trajectory frames are generated.

cs.CE↗