arXiv subjects

Shanshan Wang

Publications and source records attributed to Shanshan Wang.

At least 163 records · Page 9Linked to original sources

arXiv2022

BP-Triplet Net for Unsupervised Domain Adaptation: A Bayesian Perspective

Triplet loss, one of the deep metric learning (DML) methods, is to learn the embeddings where examples from the same class are closer than examples from different classes. Motivated by DML, we propose an effective BP-Triplet Loss for unsupervised domain adaption (UDA) from the perspective of Bayesian learning and we name the model as BP-Triplet Net. In previous metric learning based methods for UDA, sample pairs across domains are treated equally, which is not appropriate due to the domain bias. In our work, considering the different importance of pair-wise samples for both feature learning and domain alignment, we deduce our BP-Triplet loss for effective UDA from the perspective of Bayesian learning. Our BP-Triplet loss adjusts the weights of pair-wise samples in intra domain and inter domain. Especially, it can self attend to the hard pairs (including hard positive pair and hard negative pair). Together with the commonly used adversarial loss for domain alignment, the quality of target pseudo labels is progressively improved. Our method achieved low joint error of the ideal source and target hypothesis. The expected target error can then be upper bounded following Ben-David s theorem. Comprehensive evaluations on five benchmark datasets, handwritten digits, Office31, ImageCLEF-DA, Office-Home and VisDA-2017 demonstrate the effectiveness of the proposed approach for UDA.

cs.CV↗

arXiv2022

Effective Lifetime of Non-Equilibrium Carriers in Semiconductors from Non-Adiabatic Molecular Dynamics Simulations

The lifetime of non-equilibrium electrons and holes in semiconductors is crucial for solar cell and optoelectronic applications. Non-adiabatic molecular dynamics (NAMD) simulations based on time-dependent density functional theory (TDDFT) are widely used to study excited-state carrier dynamics. However, the calculated carrier lifetimes are often different from experimental results by orders of magnitude. In this work, by revisiting the definition of carrier lifetime and considering different recombination mechanisms, we report a systematic procedure for calculating the effective carrier lifetime in realistic semiconductor crystals that can be compared directly to experimental measurements. The procedure shows that considering all recombination mechanisms and using reasonable densities of carriers and defects are crucial in calculating the effective lifetime. When NAMD simulations consider only Shockey-Read-Hall (SRH) defect-assisted and band-to-band non-radiative recombination while neglect band-to-band radiative recombination, and the densities of non-equilibrium carriers and defects in supercell simulations are much higher than those in realistic semiconductors under solar illumination, the calculated lifetimes are ineffective and thus differ from experiments. Using our procedure, the calculated effective lifetime of the halide perovskite CH3NH3PbI3 agrees with experiments. It is mainly determined by band-to-band radiative and defect-assisted non-radiative recombination, while band-to-band non-radiative recombination is negligible. These results indicate that it is possible to calculate carrier lifetimes accurately based on NAMD simulations, but the directly calculated values should be converted to effective lifetimes for comparison to experiments. The revised procedure can be widely applied in future carrier lifetime simulations.

cond-mat.mtrl-sci↗

arXiv2022

Universal Generative Modeling for Calibration-free Parallel Mr Imaging

The integration of compressed sensing and parallel imaging (CS-PI) provides a robust mechanism for accelerating MRI acquisitions. However, most such strategies require the explicit formation of either coil sensitivity profiles or a cross-coil correlation operator, and as a result reconstruction corresponds to solving a challenging bilinear optimization problem. In this work, we present an unsupervised deep learning framework for calibration-free parallel MRI, coined universal generative modeling for parallel imaging (UGM-PI). More precisely, we make use of the merits of both wavelet transform and the adaptive iteration strategy in a unified framework. We train a powerful noise conditional score network by forming wavelet tensor as the network input at the training phase. Experimental results on both physical phantom and in vivo datasets implied that the proposed method is comparable and even superior to state-of-the-art CS-PI reconstruction approaches.

cs.CV↗

arXiv2022

Expert Knowledge-guided Geometric Representation Learning for Magnetic Resonance Imaging-based Glioma Grading

Radiomics and deep learning have shown high popularity in automatic glioma grading. Radiomics can extract hand-crafted features that quantitatively describe the expert knowledge of glioma grades, and deep learning is powerful in extracting a large number of high-throughput features that facilitate the final classification. However, the performance of existing methods can still be improved as their complementary strengths have not been sufficiently investigated and integrated. Furthermore, lesion maps are usually needed for the final prediction at the testing phase, which is very troublesome. In this paper, we propose an expert knowledge-guided geometric representation learning (ENROL) framework . Geometric manifolds of hand-crafted features and learned features are constructed to mine the implicit relationship between deep learning and radiomics, and therefore to dig mutual consent and essential representation for the glioma grades. With a specially designed manifold discrepancy measurement, the grading model can exploit the input image data and expert knowledge more effectively in the training phase and get rid of the requirement of lesion segmentation maps at the testing phase. The proposed framework is flexible regarding deep learning architectures to be utilized. Three different architectures have been evaluated and five models have been compared, which show that our framework can always generate promising results.

eess.IV↗

arXiv2022

DASP: Defect and Dopant ab-initio Simulation Package

In order to perform automated calculations of defect and dopant properties in semiconductors and insulators, we developed a software package, Defect and Dopant ab-initio Simulation Package (DASP), which is composed of four modules for calculating: (i) elemental chemical potentials, (ii) defect (dopant) formation energies and transition energy levels, (iii) defect and carrier densities and (iv) carrier dynamics properties of high-density defects. DASP uses the materials genome database for quick determination of competing secondary phases and calculation of the energy above convex hull when calculating the elemental chemical potential that stabilizes compound semiconductors, so it can perform high-throughput prediction of thermodynamic stability of multinary compounds. DASP calls the ab-initio softwares to perform the total energy, structural relaxation and electronic structure calculations of the defect supercells with different structure configurations and charge states, based on which the defect formation energies and transition energy levels are calculated and the corrections for electrostatic potential alignment and image charge interaction can be included. Then DASP can calculate the equilibrium densities of defects and electron and hole carriers as well as the Fermi level in semiconductors under different chemical potential conditions and different growth/working temperature. For high-density defects, DASP can calculate the carrier dynamics properties such as the photoluminescence (PL) spectrum, defect-related radiative and non-radiative carrier capture cross sections, and recombination lifetime of non-equilibrium carriers.

cond-mat.mtrl-sci↗

arXiv2022

Collective behavior in the North Rhine-Westphalia motorway network

To understand the dynamics on complex networks, measurement of correlations is indispensable. In a motorway network, it is not sufficient to collect information on fluxes and velocities on all individual links, i.e. parts of the freeways between ramps and highway crosses. The interdependencies and mutual connections are also of considerable interest. We analyze correlations in the complete motorway network in North Rhine-Westphalia, the most populous state in Germany. We view the motorway network as a complex system consisting of road sections which interact via the motion of vehicles, implying structures in the corresponding correlation matrices. In particular, we focus on collective behavior, i.e. coherent motion in the whole network or in large parts of it. To this end, we study the eigenvalue and eigenvector statistics and identify significant sections in the motorway network. We find collective behavior in these significant sections and further explore its causes. We show that collectivity throughout the network cannot directly be related to the traffic states (free, synchronous and congested) in Kerner's three-phase theory. Hence, the degree of collectivity provides a new, complementary observable to characterize the motorway network.

physics.soc-ph↗

arXiv2021

Radiomic biomarker extracted from PI-RADS 3 patients support more eìcient and robust prostate cancer diagnosis: a multi-center study

Prostate Imaging Reporting and Data System (PI-RADS) based on multi-parametric MRI classiêes patients into 5 categories (PI-RADS 1-5) for routine clinical diagnosis guidance. However, there is no consensus on whether PI-RADS 3 patients should go through biopsies. Mining features from these hard samples (HS) is meaningful for physicians to achieve accurate diagnoses. Currently, the mining of HS biomarkers is insuìcient, and the eéectiveness and robustness of HS biomarkers for prostate cancer diagnosis have not been explored. In this study, biomarkers from diéerent data distributions are constructed. Results show that HS biomarkers can achieve better performances in diéerent data distributions.

eess.IV↗

arXiv2021

Ferroelectricity in strained Hf$_2$CF$_2$ monolayer

Low dimensional ferroelectrics are highly desired for applications and full of exotic physics. Here a functionalized MXene Hf$_2$CF$_2$ monolayer is theoretically studied, which manifests a nonpolar to polar transition upon moderate biaxial compressive strain. Accompanying this structural transition, a metal-semiconductor transition occurs. The in-plane shift of unilateral fluorine layer leads to a polarization pointing out-of-plane. Such ferroelectricity is unconventional, similar to the recently-proposed interlayer-sliding ferroelectricity but not identical. Due to its specific hexapetalous potential energy profile, the possible ferroelectric switching paths and domain walls are nontrivial, which are mediated via the metallic paraelectric state. In this sense, the metallic walls can be manipulated by reshaping the ferroelectric domains.

cond-mat.mtrl-sci↗

arXiv2021

Self-Supervised Learning for MRI Reconstruction with a Parallel Network Training Framework

Image reconstruction from undersampled k-space data plays an important role in accelerating the acquisition of MR data, and a lot of deep learning-based methods have been exploited recently. Despite the achieved inspiring results, the optimization of these methods commonly relies on the fully-sampled reference data, which are time-consuming and difficult to collect. To address this issue, we propose a novel self-supervised learning method. Specifically, during model optimization, two subsets are constructed by randomly selecting part of k-space data from the undersampled data and then fed into two parallel reconstruction networks to perform information recovery. Two reconstruction losses are defined on all the scanned data points to enhance the network's capability of recovering the frequency information. Meanwhile, to constrain the learned unscanned data points of the network, a difference loss is designed to enforce consistency between the two parallel networks. In this way, the reconstruction model can be properly trained with only the undersampled data. During the model evaluation, the undersampled data are treated as the inputs and either of the two trained networks is expected to reconstruct the high-quality results. The proposed method is flexible and can be employed in any existing deep learning-based method. The effectiveness of the method is evaluated on an open brain MRI dataset. Experimental results demonstrate that the proposed self-supervised method can achieve competitive reconstruction performance compared to the corresponding supervised learning method at high acceleration rates (4 and 8). The code is publicly available at \url{https://github.com/chenhu96/Self-Supervised-MRI-Reconstruction}.

cs.CV↗

arXiv2021

Group Shift Pointwise Convolution for Volumetric Medical Image Segmentation

Recent studies have witnessed the effectiveness of 3D convolutions on segmenting volumetric medical images. Compared with the 2D counterparts, 3D convolutions can capture the spatial context in three dimensions. Nevertheless, models employing 3D convolutions introduce more trainable parameters and are more computationally complex, which may lead easily to model overfitting especially for medical applications with limited available training data. This paper aims to improve the effectiveness and efficiency of 3D convolutions by introducing a novel Group Shift Pointwise Convolution (GSP-Conv). GSP-Conv simplifies 3D convolutions into pointwise ones with 1x1x1 kernels, which dramatically reduces the number of model parameters and FLOPs (e.g. 27x fewer than 3D convolutions with 3x3x3 kernels). Naïve pointwise convolutions with limited receptive fields cannot make full use of the spatial image context. To address this problem, we propose a parameter-free operation, Group Shift (GS), which shifts the feature maps along with different spatial directions in an elegant way. With GS, pointwise convolutions can access features from different spatial locations, and the limited receptive fields of pointwise convolutions can be compensated. We evaluate the proposed methods on two datasets, PROMISE12 and BraTS18. Results show that our method, with substantially decreased model complexity, achieves comparable or even better performance than models employing 3D convolutions.

eess.IV↗

arXiv2021

A Novel Hybrid Convolutional Neural Network for Accurate Organ Segmentation in 3D Head and Neck CT Images

Radiation therapy (RT) is widely employed in the clinic for the treatment of head and neck (HaN) cancers. An essential step of RT planning is the accurate segmentation of various organs-at-risks (OARs) in HaN CT images. Nevertheless, segmenting OARs manually is time-consuming, tedious, and error-prone considering that typical HaN CT images contain tens to hundreds of slices. Automated segmentation algorithms are urgently required. Recently, convolutional neural networks (CNNs) have been extensively investigated on this task. Particularly, 3D CNNs are frequently adopted to process 3D HaN CT images. There are two issues with naïve 3D CNNs. First, the depth resolution of 3D CT images is usually several times lower than the in-plane resolution. Direct employment of 3D CNNs without distinguishing this difference can lead to the extraction of distorted image features and influence the final segmentation performance. Second, a severe class imbalance problem exists, and large organs can be orders of times larger than small organs. It is difficult to simultaneously achieve accurate segmentation for all the organs. To address these issues, we propose a novel hybrid CNN that fuses 2D and 3D convolutions to combat the different spatial resolutions and extract effective edge and semantic features from 3D HaN CT images. To accommodate large and small organs, our final model, named OrganNet2.5D, consists of only two instead of the classic four downsampling operations, and hybrid dilated convolutions are introduced to maintain the respective field. Experiments on the MICCAI 2015 challenge dataset demonstrate that OrganNet2.5D achieves promising performance compared to state-of-the-art methods.

eess.IV↗

arXiv2021

Annotation-efficient deep learning for automatic medical image segmentation

Automatic medical image segmentation plays a critical role in scientific research and medical care. Existing high-performance deep learning methods typically rely on large training datasets with high-quality manual annotations, which are difficult to obtain in many clinical applications. Here, we introduce Annotation-effIcient Deep lEarning (AIDE), an open-source framework to handle imperfect training datasets. Methodological analyses and empirical evaluations are conducted, and we demonstrate that AIDE surpasses conventional fully-supervised models by presenting better performance on open datasets possessing scarce or noisy annotations. We further test AIDE in a real-life case study for breast tumor segmentation. Three datasets containing 11,852 breast images from three medical centers are employed, and AIDE, utilizing 10% training annotations, consistently produces segmentation maps comparable to those generated by fully-supervised counterparts or provided by independent radiologists. The 10-fold enhanced efficiency in utilizing expert labels has the potential to promote a wide range of biomedical applications.

eess.IV↗

arXiv2021

MRI Reconstruction Using Deep Energy-Based Model

Purpose: Although recent deep energy-based generative models (EBMs) have shown encouraging results in many image generation tasks, how to take advantage of the self-adversarial cogitation in deep EBMs to boost the performance of Magnetic Resonance Imaging (MRI) reconstruction is still desired. Methods: With the successful application of deep learning in a wide range of MRI reconstruction, a line of emerging research involves formulating an optimization-based reconstruction method in the space of a generative model. Leveraging this, a novel regularization strategy is introduced in this article which takes advantage of self-adversarial cogitation of the deep energy-based model. More precisely, we advocate for alternative learning a more powerful energy-based model with maximum likelihood estimation to obtain the deep energy-based information, represented as image prior. Simultaneously, implicit inference with Langevin dynamics is a unique property of re-construction. In contrast to other generative models for reconstruction, the proposed method utilizes deep energy-based information as the image prior in reconstruction to improve the quality of image. Results: Experiment results that imply the proposed technique can obtain remarkable performance in terms of high reconstruction accuracy that is competitive with state-of-the-art methods, and does not suffer from mode collapse. Conclusion: Algorithmically, an iterative approach was presented to strengthen EBM training with the gradient of energy network. The robustness and the reproducibility of the algorithm were also experimentally validated. More importantly, the proposed reconstruction framework can be generalized for most MRI reconstruction scenarios.

eess.IV↗

arXiv2021

Audio-visual scene classification: analysis of DCASE 2021 Challenge submissions

This paper presents the details of the Audio-Visual Scene Classification task in the DCASE 2021 Challenge (Task 1 Subtask B). The task is concerned with classification using audio and video modalities, using a dataset of synchronized recordings. This task has attracted 43 submissions from 13 different teams around the world. Among all submissions, more than half of the submitted systems have better performance than the baseline. The common techniques among the top systems are the usage of large pretrained models such as ResNet or EfficientNet which are trained for the task-specific problem. Fine-tuning, transfer learning, and data augmentation techniques are also employed to boost the performance. More importantly, multi-modal methods using both audio and video are employed by all the top 5 teams. The best system among all achieved a logloss of 0.195 and accuracy of 93.8%, compared to the baseline system with logloss of 0.662 and accuracy of 77.1%.

eess.AS↗

arXiv2021

Blind Image Quality Assessment for MRI with A Deep Three-dimensional content-adaptive Hyper-Network

Image Quality Assessment (IQA) is of great value in the workflow of Magnetic Resonance Imaging (MRI)-based analysis. Blind IQA (BIQA) methods are especially required since high-quality reference MRI images are usually not available. Recently, many efforts have been devoted to developing deep learning-based BIQA approaches. However, the performance of these methods is limited due to the utilization of simple content-non-adaptive network parameters and the waste of the important 3D spatial information of the medical images. To address these issues, we design a 3D content-adaptive hyper-network for MRI BIQA. The overall 3D configuration enables the exploration of comprehensive 3D spatial information from MRI images, while the developed content-adaptive hyper-network contributes to the self-adaptive capacity of network parameters and thus, facilitates better BIQA performance. The effectiveness of the proposed method is extensively evaluated on the open dataset, MRIQC. Promising performance is achieved compared with the corresponding baseline and 4 state-of-the-art BIQA methods. We make our code available at \url{https://git.openi.org.cn/SIAT_Wangshanshan/HyS-Net}.

eess.IV↗

arXiv2021

Few-Shot Electronic Health Record Coding through Graph Contrastive Learning

Electronic health record (EHR) coding is the task of assigning ICD codes to each EHR. Most previous studies either only focus on the frequent ICD codes or treat rare and frequent ICD codes in the same way. These methods perform well on frequent ICD codes but due to the extremely unbalanced distribution of ICD codes, the performance on rare ones is far from satisfactory. We seek to improve the performance for both frequent and rare ICD codes by using a contrastive graph-based EHR coding framework, CoGraph, which re-casts EHR coding as a few-shot learning task. First, we construct a heterogeneous EHR word-entity (HEWE) graph for each EHR, where the words and entities extracted from an EHR serve as nodes and the relations between them serve as edges. Then, CoGraph learns similarities and dissimilarities between HEWE graphs from different ICD codes so that information can be transferred among them. In a few-shot learning scenario, the model only has access to frequent ICD codes during training, which might force it to encode features that are useful for frequent ICD codes only. To mitigate this risk, CoGraph devises two graph contrastive learning schemes, GSCL and GECL, that exploit the HEWE graph structures so as to encode transferable features. GSCL utilizes the intra-correlation of different sub-graphs sampled from HEWE graphs while GECL exploits the inter-correlation among HEWE graphs at different clinical stages. Experiments on the MIMIC-III benchmark dataset show that CoGraph significantly outperforms state-of-the-art methods on EHR coding, not only on frequent ICD codes, but also on rare codes, in terms of several evaluation indicators. On frequent ICD codes, GSCL and GECL improve the classification accuracy and F1 by 1.31% and 0.61%, respectively, and on rare ICD codes CoGraph has more obvious improvements by 2.12% and 2.95%.

cs.AI↗

arXiv2021

Deep neural network Based Low-latency Speech Separation with Asymmetric analysis-Synthesis Window Pair

Time-frequency masking or spectrum prediction computed via short symmetric windows are commonly used in low-latency deep neural network (DNN) based source separation. In this paper, we propose the usage of an asymmetric analysis-synthesis window pair which allows for training with targets with better frequency resolution, while retaining the low-latency during inference suitable for real-time speech enhancement or assisted hearing applications. In order to assess our approach across various model types and datasets, we evaluate it with both speaker-independent deep clustering (DC) model and a speaker-dependent mask inference (MI) model. We report an improvement in separation performance of up to 1.5 dB in terms of source-to-distortion ratio (SDR) while maintaining an algorithmic latency of 8 ms.

eess.AS↗

arXiv2021

Dimension reduction of open-high-low-close data in candlestick chart based on pseudo-PCA

The (open-high-low-close) OHLC data is the most common data form in the field of finance and the investigate object of various technical analysis. With increasing features of OHLC data being collected, the issue of extracting their useful information in a comprehensible way for visualization and easy interpretation must be resolved. The inherent constraints of OHLC data also pose a challenge for this issue. This paper proposes a novel approach to characterize the features of OHLC data in a dataset and then performs dimension reduction, which integrates the feature information extraction method and principal component analysis. We refer to it as the pseudo-PCA method. Specifically, we first propose a new way to represent the OHLC data, which will free the inherent constraints and provide convenience for further analysis. Moreover, there is a one-to-one match between the original OHLC data and its feature-based representations, which means that the analysis of the feature-based data can be reversed to the original OHLC data. Next, we develop the pseudo-PCA procedure for OHLC data, which can effectively identify important information and perform dimension reduction. Finally, the effectiveness and interpretability of the proposed method are investigated through finite simulations and the spot data of China's agricultural product market.

econ.EM↗