A5100556775- Image and Signal Denoising Methods
- Online Learning and Analytics
- 3D Surveying and Cultural Heritage
- Neural Networks and Applications
- Imbalanced Data Classification Techniques
- Speech and Audio Processing
- Machine Learning and ELM
- Face and Expression Recognition
- 3D Shape Modeling and Analysis
- Web Data Mining and Analysis
- Sparse and Compressive Sensing Techniques
- Fault Detection and Control Systems
- Stochastic Gradient Optimization Techniques
- Educational Assessment and Pedagogy
- Remote-Sensing Image Classification
- Robotics and Sensor-Based Localization
- Remote Sensing and LiDAR Applications
- Video Surveillance and Tracking Methods
- Target Tracking and Data Fusion in Sensor Networks
- Gait Recognition and Analysis
- Adversarial Robustness in Machine Learning
- Gaussian Processes and Bayesian Inference
- Advanced Computing and Algorithms
- Digital Media Forensic Detection
- Dengue and Mosquito Control Research
Nantong University
2024-2025
University of Missouri–Kansas City
2017-2024
Shanghai First People's Hospital
2024
Shanghai Jiao Tong University
2016-2024
Nanjing University
2024
Henan Provincial People's Hospital
2024
Xi'an Jiaotong University
2023-2024
Zhengzhou University
2024
Dalian University of Technology
2023
University College London
2021
In this paper, we propose an efficient and effective framework to fuse hyperspectral Light Detection And Ranging (LiDAR) data using two coupled convolutional neural networks (CNNs). One CNN is designed learn spectral-spatial features from data, the other one used capture elevation information LiDAR data. Both of them consist three layers, last layers are together via a parameter sharing strategy. fusion phase, feature-level decision-level methods simultaneously integrate these heterogeneous...
For chroma intra prediction, previous methods exemplified by the Linear Model method (LM) usually assume a linear correlation between luma and components in coding block. This assumption is inaccurate for complex image content or large blocks, restricts prediction accuracy. In this paper, we propose exploiting both spatial cross-channel correlations using hybrid neural network. Specifically, utilize convolutional network to extract features from reconstructed samples of current block, as...
Obesity is the main risk factor leading to development of various respiratory diseases, such as asthma and pulmonary hypertension. Pulmonary microvascular endothelial cells (PMVECs) play a significant role in lung diseases. Aconitate decarboxylase 1 (Acod1) mediates production itaconate, Acod1/itaconate axis has been reported protective multiple However, roles PMVECs obese mice are still unclear. mRNA-seq was performed identify differentially expressed genes (DEGs) between high-fat diet...
Emotion recognition from facial expressions is tremendously useful, especially when coupled with smart devices and wireless multimedia applications. However, the inadequate network bandwidth often limits spatial resolution of transmitted video, which will heavily degrade reliability. We develop a novel framework to achieve robust emotion low bit rate video. While video frames are downsampled at encoder side, decoder embedded deep model for joint super-resolution (SR) recognition. Notably, we...
One of the radar high resolution range profile (HRRP) target recognition issues is existence noise interference, especially for ground target. The performance traditional shallow methods degrades as suffering from limited capability extracting robust and discriminative features. In this paper, a novel deep neural network called stacked denoising contractive auto-encoder (SDCAE) designed millimeter wave HRRP recognition. To enhance learning structure correlations corrupted data, by combining...
In this article, the issue of joint state and fault estimation is ironed out for delayed state-saturated systems subject to energy harvesting sensors. Under effect harvesting, sensors can harvest from external environment consume an amount when transmitting measurements estimator. The occurrence probability measurement loss computed at each instant according distribution mechanism. main objective addressed problem construct a estimator where error covariance ensured in some certain sense...
Attention-deficit/hyperactivity disorder (ADHD) is one of the most common and controversial diseases in paediatric psychiatry. Recently, computer-aided diagnosis methods become increasingly popular clinical ADHD. In this paper, we introduced latest powerful method—deep convolutional neural networks (CNNs). Some data augmentation CNN transfer learning technique were used to address application problem deep CNNs ADHD classification task, given limited annotated data. addition, previously...
Modern machine learning often operates in the regime where number of parameters is much higher than data points, with zero training loss and yet good generalization, thereby contradicting classical bias-variance trade-off. This \textit{benign overfitting} phenomenon has recently been characterized using so called \textit{double descent} curves risk undergoes another descent (in addition to U-shaped curve when small) as we increase beyond a certain threshold. In this paper, examine conditions...
Modern machine learning models often employ a huge number of parameters and are typically optimized to have zero training loss; yet surprisingly, they possess near-optimal prediction performance, contradicting classical theory. We examine how these benign overfitting phenomena occur in two-layer neural network setting where sample covariates corrupted with noise. address the high dimensional regime, data dimension $d$ grows $n$ points. Our analysis combines an upper bound on bias matching...
Mosquitoes are of great concern for occasionally carrying noxious diseases (dengue, malaria, zika, and yellow fever). To control mosquitoes, it is very crucial to effectively monitor their behavioral trends presence. Traditional mosquito repellent works by heating small pads soaked in repellant, which then diffuses a protected area around you, alternative spraying yourself with insecticide. But they have limitations, including the range, turning them on manually, waiting protection kick when...
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Abstract Background Anthracycline and taxane-based chemotherapy are the corner stone of neoadjuvant therapy in early breast cancer. Anthracycline-free regimens have been addressed widely because cardiac toxicity especially HER2-positive cancer when combining trastuzumab pertuzumab. Aim this trial was to assess efficacy safety nab-paclitaxel combined with carboplatin non-luminal Methods The Neopath a prospective, single-arm phase II conducted Comprehensive Breast Health Center, Ruijin...
Abstract Triple-negative breast cancer (TNBC) is one of the most aggressive subtypes cancer, with high malignancy and poor outcome. Resistance to chemotherapy leads treatment failure. How overcome drug resistance key point in TNBC treatment. Previously, we initiated a prospective multicenter single-arm NeoPath clinical study (ClinicalTrial.gov identifier: NCT03907800) assess efficacy safety nab-paclitaxel combined carboplatin neoadjuvant therapy triple negative HER2-positive early cancer. A...
We study theoretical properties of a broad class regularized algorithms with vector-valued output. These spectral include kernel ridge regression, principal component various implementations gradient descent and many more. Our contributions are twofold. First, we rigorously confirm the so-called saturation effect for regression output by deriving novel lower bound on learning rates; this is shown to be suboptimal when smoothness function exceeds certain level. Second, present upper finite...
This paper deals with the recursive state estimation issue for mobile robot localization under a dynamic event-based mechanism. To enhance utilization of communication resources, transmission protocol is utilized to reduce unnecessary measurement transmissions by introducing an auxiliary dynamical variable adjust threshold parameters. The primary objective this develop scheme problem in presence impact mechanism such that upper bound on error covariance firstly guaranteed using mathematical...