arXiv subjects

Hui Shen

Publications and source records attributed to Hui Shen.

At least 91 records · Page 5Linked to original sources

arXiv2023

CLCLSA: Cross-omics Linked embedding with Contrastive Learning and Self Attention for multi-omics integration with incomplete multi-omics data

Integration of heterogeneous and high-dimensional multi-omics data is becoming increasingly important in understanding genetic data. Each omics technique only provides a limited view of the underlying biological process and integrating heterogeneous omics layers simultaneously would lead to a more comprehensive and detailed understanding of diseases and phenotypes. However, one obstacle faced when performing multi-omics data integration is the existence of unpaired multi-omics data due to instrument sensitivity and cost. Studies may fail if certain aspects of the subjects are missing or incomplete. In this paper, we propose a deep learning method for multi-omics integration with incomplete data by Cross-omics Linked unified embedding with Contrastive Learning and Self Attention (CLCLSA). Utilizing complete multi-omics data as supervision, the model employs cross-omics autoencoders to learn the feature representation across different types of biological data. The multi-omics contrastive learning, which is used to maximize the mutual information between different types of omics, is employed before latent feature concatenation. In addition, the feature-level self-attention and omics-level self-attention are employed to dynamically identify the most informative features for multi-omics data integration. Extensive experiments were conducted on four public multi-omics datasets. The experimental results indicated that the proposed CLCLSA outperformed the state-of-the-art approaches for multi-omics data classification using incomplete multi-omics data.

cs.LG↗

arXiv2023

Multi-view information fusion using multi-view variational autoencoders to predict proximal femoral strength

The aim of this paper is to design a deep learning-based model to predict proximal femoral strength using multi-view information fusion. Method: We developed new models using multi-view variational autoencoder (MVAE) for feature representation learning and a product of expert (PoE) model for multi-view information fusion. We applied the proposed models to an in-house Louisiana Osteoporosis Study (LOS) cohort with 931 male subjects, including 345 African Americans and 586 Caucasians. With an analytical solution of the product of Gaussian distribution, we adopted variational inference to train the designed MVAE-PoE model to perform common latent feature extraction. We performed genome-wide association studies (GWAS) to select 256 genetic variants with the lowest p-values for each proximal femoral strength and integrated whole genome sequence (WGS) features and DXA-derived imaging features to predict proximal femoral strength. Results: The best prediction model for fall fracture load was acquired by integrating WGS features and DXA-derived imaging features. The designed models achieved the mean absolute percentage error of 18.04%, 6.84% and 7.95% for predicting proximal femoral fracture loads using linear models of fall loading, nonlinear models of fall loading, and nonlinear models of stance loading, respectively. Compared to existing multi-view information fusion methods, the proposed MVAE-PoE achieved the best performance. Conclusion: The proposed models are capable of predicting proximal femoral strength using WGS features and DXA-derived imaging features. Though this tool is not a substitute for FEA using QCT images, it would make improved assessment of hip fracture risk more widely available while avoiding the increased radiation dosage and clinical costs from QCT.

cs.LG↗

arXiv2023

Dependence of ocean surface filaments on wind speed: An observational study of North Atlantic right whale habitat

Coherent filaments at the ocean surface often appear to be transient watermass boundaries, where currents converge, surfactants accumulate, and frontal structure at depth can possibly delineate enhanced biological activity in the upper water column. Spaceborne synthetic aperture radar (SAR) permits filaments to be observed at O[1-km] resolution, but extensive coherent structures are more apparent in weaker winds. A wind speed adjustment is proposed for filaments (i.e., contiguous SAR contrasts) of at least 10 km in length. Measures of dependence (distance correlation and the linear and nonlinear components of Pearson correlation) are examined to identify a broad peak in the relationship between filament contrast and weak or moderate values of surface wind speed, where a variable wind speed exponent is employed to maximize these measures. Three locations of recent North Atlantic right whale (Eubalaena glacialis) sightings in the Gulf of St. Lawrence are sampled between 2008 and 2020 by 324 Radarsat-2 SAR scenes and 10-m wind speed from the ERA5 reanalysis. The inverse relationship between SAR contrast magnitude and wind speed is quantified, and a reduced correlation is obtained for all three domains when SAR contrast is weighted by wind speed to the power of 0.8. A more uniform emphasis on ocean surface structure within a SAR scene, or across multiple scenes, can thus be considered in the search for prey aggregations of the North Atlantic right whale.

stat.AP↗

arXiv2022

A robust kernel machine regression towards biomarker selection in multi-omics datasets of osteoporosis for drug discovery

Many statistical machine approaches could ultimately highlight novel features of the etiology of complex diseases by analyzing multi-omics data. However, they are sensitive to some deviations in distribution when the observed samples are potentially contaminated with adversarial corrupted outliers (e.g., a fictional data distribution). Likewise, statistical advances lag in supporting comprehensive data-driven analyses of complex multi-omics data integration. We propose a novel non-linear M-estimator-based approach, "robust kernel machine regression (RobKMR)," to improve the robustness of statistical machine regression and the diversity of fictional data to examine the higher-order composite effect of multi-omics datasets. We address a robust kernel-centered Gram matrix to estimate the model parameters accurately. We also propose a robust score test to assess the marginal and joint Hadamard product of features from multi-omics data. We apply our proposed approach to a multi-omics dataset of osteoporosis (OP) from Caucasian females. Experiments demonstrate that the proposed approach effectively identifies the inter-related risk factors of OP. With solid evidence (p-value = 0.00001), biological validations, network-based analysis, causal inference, and drug repurposing, the selected three triplets ((DKK1, SMTN, DRGX), (MTND5, FASTKD2, CSMD3), (MTND5, COG3, CSMD3)) are significant biomarkers and directly relate to BMD. Overall, the top three selected genes (DKK1, MTND5, FASTKD2) and one gene (SIDT1 at p-value= 0.001) significantly bond with four drugs- Tacrolimus, Ibandronate, Alendronate, and Bazedoxifene out of 30 candidates for drug repurposing in OP. Further, the proposed approach can be applied to any disease model where multi-omics datasets are available.

stat.ML↗

arXiv2021

Frequency-specific segregation and integration of human cerebral cortex: an intrinsic functional atlas

The frequency-specific coupling mechanism of the functional human brain networks underpins its complex cognitive and behavioral functions. Nevertheless, it is not well unveiled what are the frequency-specific subdivisions and network topologies of the human brain. In this study, we estimated functional connectivity of the human cerebral cortex using spectral connection, and conducted frequency-specific parcellation using eigen-clustering and gradient-based methods, and then explored their topological structures. 7T fMRI data of 184 subjects in the HCP dataset were used for parcellation and exploring the topological properties of the functional networks, and 3T fMRI data of another 890 subjects were used to confirm the stability of the frequency-specific topologies. Seven to ten functional networks were stably integrated by two to four dissociable hub categories at specific frequencies, and we proposed an intrinsic functional atlas containing 456 parcels according to the parcellations across frequencies. The results revealed that the functional networks contained stable frequency-specific topologies, which may imply more abundant roles of the functional units and more complex interactions among them.

q-bio.NC↗

arXiv2020

PP-YOLO: An Effective and Efficient Implementation of Object Detector

Object detection is one of the most important areas in computer vision, which plays a key role in various practical scenarios. Due to limitation of hardware, it is often necessary to sacrifice accuracy to ensure the infer speed of the detector in practice. Therefore, the balance between effectiveness and efficiency of object detector must be considered. The goal of this paper is to implement an object detector with relatively balanced effectiveness and efficiency that can be directly applied in actual application scenarios, rather than propose a novel detection model. Considering that YOLOv3 has been widely used in practice, we develop a new object detector based on YOLOv3. We mainly try to combine various existing tricks that almost not increase the number of model parameters and FLOPs, to achieve the goal of improving the accuracy of detector as much as possible while ensuring that the speed is almost unchanged. Since all experiments in this paper are conducted based on PaddlePaddle, we call it PP-YOLO. By combining multiple tricks, PP-YOLO can achieve a better balance between effectiveness (45.2% mAP) and efficiency (72.9 FPS), surpassing the existing state-of-the-art detectors such as EfficientDet and YOLOv4.Source code is at https://github.com/PaddlePaddle/PaddleDetection.

cs.CV↗

arXiv2020

A Deep Learning-Based Method for Automatic Segmentation of Proximal Femur from Quantitative Computed Tomography Images

Purpose: Proximal femur image analyses based on quantitative computed tomography (QCT) provide a method to quantify the bone density and evaluate osteoporosis and risk of fracture. We aim to develop a deep-learning-based method for automatic proximal femur segmentation. Methods and Materials: We developed a 3D image segmentation method based on V-Net, an end-to-end fully convolutional neural network (CNN), to extract the proximal femur QCT images automatically. The proposed V-net methodology adopts a compound loss function, which includes a Dice loss and a L2 regularizer. We performed experiments to evaluate the effectiveness of the proposed segmentation method. In the experiments, a QCT dataset which included 397 QCT subjects was used. For the QCT image of each subject, the ground truth for the proximal femur was delineated by a well-trained scientist. During the experiments for the entire cohort then for male and female subjects separately, 90% of the subjects were used in 10-fold cross-validation for training and internal validation, and to select the optimal parameters of the proposed models; the rest of the subjects were used to evaluate the performance of models. Results: Visual comparison demonstrated high agreement between the model prediction and ground truth contours of the proximal femur portion of the QCT images. In the entire cohort, the proposed model achieved a Dice score of 0.9815, a sensitivity of 0.9852 and a specificity of 0.9992. In addition, an R2 score of 0.9956 (p<0.001) was obtained when comparing the volumes measured by our model prediction with the ground truth. Conclusion: This method shows a great promise for clinical application to QCT and QCT-based finite element analysis of the proximal femur for evaluating osteoporosis and hip fracture risk.

physics.med-ph↗

arXiv2020

A generalized kernel machine approach to identify higher-order composite effects in multi-view datasets

In recent years, a comprehensive study of multi-view datasets (e.g., multi-omics and imaging scans) has been a focus and forefront in biomedical research. State-of-the-art biomedical technologies are enabling us to collect multi-view biomedical datasets for the study of complex diseases. While all the views of data tend to explore complementary information of a disease, multi-view data analysis with complex interactions is challenging for a deeper and holistic understanding of biological systems. In this paper, we propose a novel generalized kernel machine approach to identify higher-order composite effects in multi-view biomedical datasets. This generalized semi-parametric (a mixed-effect linear model) approach includes the marginal and joint Hadamard product of features from different views of data. The proposed kernel machine approach considers multi-view data as predictor variables to allow more thorough and comprehensive modeling of a complex trait. The proposed method can be applied to the study of any disease model, where multi-view datasets are available. We applied our approach to both synthesized datasets and real multi-view datasets from adolescence brain development and osteoporosis study, including an imaging scan dataset and five omics datasets. Our experiments demonstrate that the proposed method can effectively identify higher-order composite effects and suggest that corresponding features (genes, region of interests, and chemical taxonomies) function in a concerted effort. We show that the proposed method is more generalizable than existing ones.

stat.ML↗

arXiv2018

Solution for Large-Scale Hierarchical Object Detection Datasets with Incomplete Annotation and Data Imbalance

This report demonstrates our solution for the Open Images 2018 Challenge. Based on our detailed analysis on the Open Images Datasets (OID), it is found that there are four typical features: large-scale, hierarchical tag system, severe annotation incompleteness and data imbalance. Considering these characteristics, an amount of strategies are employed, including SNIPER, soft sampling, class-aware sampling (CAS), hierarchical non-maximum suppression (HNMS) and so on. In virtue of these effective strategies, and further using the powerful SENet154 armed with feature pyramid module and deformable ROIalign as the backbone, our best single model could achieve a mAP of 56.9%. After a further ensemble with 9 models, the final mAP is boosted to 62.2% in the public leaderboard (ranked the 2nd place) and 58.6% in the private leaderboard (ranked the 3rd place, slightly inferior to the 1st place by only 0.04 point).

cs.CV↗

arXiv2018

A Low-Latency List Successive-Cancellation Decoding Implementation for Polar Codes

Due to their provably capacity-achieving performance, polar codes have attracted a lot of research interest recently. For a good error-correcting performance, list successive-cancellation decoding (LSCD) with large list size is used to decode polar codes. However, as the complexity and delay of the list management operation rapidly increase with the list size, the overall latency of LSCD becomes large and limits the applicability of polar codes in high-throughput and latency-sensitive applications. Therefore, in this work, the low-latency implementation for LSCD with large list size is studied. Specifically, at the system level, a selective expansion method is proposed such that some of the reliable bits are not expanded to reduce the computation and latency. At the algorithmic level, a double thresholding scheme is proposed as a fast approximate-sorting method for the list management operation to reduce the LSCD latency for large list size. A VLSI architecture of the LSCD implementing the selective expansion and double thresholding scheme is then developed, and implemented using a UMC 90 nm CMOS technology. Experimental results show that, even for a large list size of 16, the proposed LSCD achieves a decoding throughput of 460 Mbps at a clock frequency of 658 MHz.

cs.IT↗

arXiv2018

A High-Throughput Architecture of List Successive Cancellation Polar Codes Decoder with Large List Size

As the first kind of forward error correction (FEC) codes that achieve channel capacity, polar codes have attracted much research interest recently. Compared with other popular FEC codes, polar codes decoded by list successive cancellation decoding (LSCD) with a large list size have better error correction performance. However, due to the serial decoding nature of LSCD and the high complexity of list management (LM), the decoding latency is high, which limits the usage of polar codes in practical applications that require low latency and high throughput. In this work, we study the high-throughput implementation of LSCD with a large list size. Specifically, at the algorithmic level, to achieve a low decoding latency with moderate hardware complexity, two decoding schemes, a multi-bit double thresholding scheme and a partial G-node look-ahead scheme, are proposed. Then, a high-throughput VLSI architecture implementing the proposed algorithms is developed with optimizations on different computation modules. From the implementation results on UMC 90 nm CMOS technology, the proposed architecture achieves decoding throughputs of 1.103 Gbps, 977 Mbps and 827 Mbps when the list sizes are 8, 16 and 32, respectively.

eess.SP↗

arXiv2016

Training Auto-encoders Effectively via Eliminating Task-irrelevant Input Variables

Auto-encoders are often used as building blocks of deep network classifier to learn feature extractors, but task-irrelevant information in the input data may lead to bad extractors and result in poor generalization performance of the network. In this paper,via dropping the task-irrelevant input variables the performance of auto-encoders can be obviously improved .Specifically, an importance-based variable selection method is proposed to aim at finding the task-irrelevant input variables and dropping them.It firstly estimates importance of each variable,and then drops the variables with importance value lower than a threshold. In order to obtain better performance, the method can be employed for each layer of stacked auto-encoders. Experimental results show that when combined with our method the stacked denoising auto-encoders achieves significantly improved performance on three challenging datasets.

cs.LG↗

arXiv2015

Capacity-Achieving Rateless Polar Codes

A rateless coding scheme transmits incrementally more and more coded bits over an unknown channel until all the information bits are decoded reliably by the receiver. We propose a new rateless coding scheme based on polar codes, and we show that this scheme is capacity-achieving, i.e. its information rate is as good as the best code specifically designed for the unknown channel. Previous rateless coding schemes are designed for specific classes of channels such as AWGN channels, binary erasure channels, etc. but the proposed rateless coding scheme is capacity-achieving for broad classes of channels as long as they are ordered via degradation. Moreover, it inherits the conceptual and computational simplicity of polar codes.

cs.IT↗

arXiv2015

Reduce the Complexity of List Decoding of Polar Codes by Tree-Pruning

Polar codes under cyclic redundancy check aided successive cancellation list (CA-SCL) decoding can outperform the turbo codes and the LDPC codes when code lengths are configured to be several kilobits. In order to reduce the decoding complexity, a novel tree-pruning scheme for the \mbox{SCL/CA-SCL} decoding algorithms is proposed in this paper. In each step of the decoding procedure, the candidate paths with metrics less than a threshold are dropped directly to avoid the unnecessary computations for the path searching on the descendant branches of them. Given a candidate path, an upper bound of the path metric of its descendants is proposed to determined whether the pruning of this candidate path would affect frame error rate (FER) performance. By utilizing this upper bounding technique and introducing a dynamic threshold, the proposed scheme deletes the redundant candidate paths as many as possible while keeping the performance deterioration in a tolerant region, thus it is much more efficient than the existing pruning scheme. With only a negligible loss of FER performance, the computational complexity of the proposed pruned decoding scheme is only about $40\%$ of the standard algorithm in the low signal-to-noise ratio (SNR) region (where the FER under CA-SCL decoding is about $0.1 \sim 0.001$), and it can be very close to that of the successive cancellation (SC) decoder in the moderate and high SNR regions.

cs.IT↗

arXiv2015

A Decision-Aided Parallel SC-List Decoder for Polar Codes

In this paper, we propose a decision-aided scheme for parallel SC-List decoding of polar codes. At the parallel SC-List decoder, each survival path is extended based on multiple information bits, therefore the number of split paths becomes very large and the sorting to find the top L paths becomes very complex. We propose a decision-aided scheme to reduce the number of split paths and thus reduce the sorting complexity.

cs.IT↗

arXiv2015

Low-latency List Decoding Of Polar Codes With Double Thresholding

For polar codes with short-to-medium code length, list successive cancellation decoding is used to achieve a good error-correcting performance. However, list pruning in the current list decoding is based on the sorting strategy and its timing complexity is high. This results in a long decoding latency for large list size. In this work, aiming at a low-latency list decoding implementation, a double thresholding algorithm is proposed for a fast list pruning. As a result, with a negligible performance degradation, the list pruning delay is greatly reduced. Based on the double thresholding, a low-latency list decoding architecture is proposed and implemented using a UMC 90nm CMOS technology. Synthesis results show that, even for a large list size of 16, the proposed low-latency architecture achieves a decoding throughput of 220 Mbps at a frequency of 641 MHz.

cs.AR↗

arXiv2014

A RM-Polar Codes

In this letter we propose a new hybrid code called "RM-Polar" codes. This new codes are constructed by combining the construction of Reed-Muller (RM) code and Polar code. It has much larger minimum Hamming distance than Polar codes, therefore it has much better error performance than Polar codes.

cs.IT↗

arXiv2013

Parallel Decoders of Polar Codes

In this letter, we propose parallel SC (Successive Cancellation) decoder and parallel SC-List decoder for polar codes. The parallel decoder is composed of M=2^m(m>=1) component decoders working in parallel and each component decoder decodes a Polar code of a block size of 1/M of the original Polar code. Therefore the parallel decoder has M times faster decoding speed. Our simulation results show that the parallel decoder has almost the same error-rate performance as the conventional non-parallel decoder.

cs.IT↗