A5062318925- AI in cancer detection
- Image Retrieval and Classification Techniques
- Medical Image Segmentation Techniques
- Advanced Graph Neural Networks
- Genomic variations and chromosomal abnormalities
- Advanced Battery Materials and Technologies
- Non-Invasive Vital Sign Monitoring
- Underwater Acoustics Research
- Chromosomal and Genetic Variations
- Image and Signal Denoising Methods
- Graph Theory and Algorithms
- Advanced Mathematical Modeling in Engineering
- Machine Learning in Materials Science
- Industrial Technology and Control Systems
- Stock Market Forecasting Methods
- Blind Source Separation Techniques
- Time Series Analysis and Forecasting
- Advancements in Battery Materials
- Radiomics and Machine Learning in Medical Imaging
- Direction-of-Arrival Estimation Techniques
- Speech and Audio Processing
- Infrared Thermography in Medicine
- Gene expression and cancer classification
- Advanced Computational Techniques and Applications
- Educational Technology and Assessment
Xi’an Jiaotong-Liverpool University
2016-2025
Peng Cheng Laboratory
2024
University of Science and Technology Beijing
2006-2023
Chongqing University
2019-2023
Northwestern Polytechnical University
2009-2023
Wuhan University
2021-2022
Applied Mathematics (United States)
2022
Chinese Academy of Medical Sciences & Peking Union Medical College
2021
Chinese Academy of Sciences
2014
Hebei University of Science and Technology
2011-2013
This paper presents the Relaxed Continuous-Time Actor-critic (RCTAC) algorithm, a method for finding nearly optimal policy nonlinear continuous-time (CT) systems with known dynamics and infinite horizon, such as path-tracking control of vehicles. RCTAC has several advantages over existing adaptive dynamic programming algorithms CT systems. It does not require “admissibility” initialized or input-affine nature controlled convergence. Instead, given any initial policy, can converge to an...
In recent years, with the increasing use of educational technology and online learning platforms, there has been a growing interest in developing intelligent systems that can automatically predict knowledge points associated questions. This paper presents novel approach for point prediction middle school mathematics The dataset used this study consists large collection 591,379 To leverage power natural language processing techniques, questions are preprocessed using tokenizer encoded into...
Depression classification often relies on multimodal features, but existing models struggle to capture the similarity between features. Moreover, social stigma surrounding depression leads limited availability of datasets, which constrains model accuracy. This study aims improve recognition methods by proposing a Multimodal Generation-Text Classification Model. The introduces Multimodal-Deep-Extract-Feature Net both long- and short-term sequential A Dual Text Contrastive Learning Module is...
Accurate prediction of a patient's length-of-stay (LOS) in the hospital enables an efficient and effective management beds. This paper studies LOS for pediatric patients with respiratory diseases using three decision tree methods: Bagging, Adaboost, Random forest. A data set 11,206 records retrieved from information system is used analysis after preprocessing transformation through computation expansion method. Two tests, namely bisection test periodic test, are designed to assess...
With the increase of operating frequency above 4 GHz, required resonator size is reduced to hundreds <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mu \text{m}^{{2}}$ </tex-math></inline-formula> , which proportional 1/ notation="LaTeX">${f}^{\,{2}}$ . In traditional, reduction will induce its parallel resonant decrease, thus resulting in decrease effective coupling coefficient (...
Electrocardiogram (ECG) signal is widely used in medical diagnosis of heart diseases. Automatic extraction relevant and reliable information from ECG signals has not been an easy task for computerized system. This study proposes to use 12-layer 1-d CNN classify 1 lead individual heartbeat into five classes The proposed method was tested on MIT/BIH arrhythmia database results were measured using positive predictive value, sensitivity F1 score. Our obtained a value 0.977, 0.976, score 0.976....
Multimodal emotion recognition techniques are increasingly essential for assessing mental states. Image-based methods, however, tend to focus predominantly on overt visual cues and often overlook subtler state changes. Psychophysiological research has demonstrated that heart rate (HR) skin temperature effective in detecting autonomic nervous system (ANS) activities, thereby revealing these subtle However, traditional HR tools generally more costly less portable, while analysis usually...
Graph similarity measurement is a fundamental task in various graph-related applications. However, recent learning-based approaches lack interpretability as they directly transform interaction information between two graphs into hidden vector, making it difficult to understand how the score derived. To address this issue, we propose an end-to-end paradigm for graph learning called Similarity Computation via Maximum Common Subgraph Inference (INFMCS), which more interpretable. Our key insight...
Karyotyping plays a crucial role in genetic disorder diagnosis. Currently requires considerable manual efforts, domain expertise and experience, is very time consuming. Automating the karyotyping process has been an important popular task. This study focuses on classification of chromosomes into 23 types, step towards fully automatic karyotyping. proposes convolutional neural network (CNN) based deep learning to automatically classify chromosomes. The proposed method was trained tested...
This paper presents an original 2D LIDAR SLAM based on point and line features to the underground tunnel environment. In our algorithm, are extracted by using Curvature detector with range information weights. The detection algorithm contains segmentation two hierarchical levels fitting. We present approach used in prototype of mining truck which achieves real-time localization mapping at a 5cm resolution. provide experimental results comparisons other approaches. show that this proposed...
For about 1800 years, tongue inspection has been one of the four major diagnostic methods in Traditional Chinese Medicine (TCM). The is believed to be able reflect health status human body. However, making an accurate diagnose with not a trivial task. It usually requires enormous training on TCM doctor before he can make reasonable diagnosis. Recently, image processing have proposed automatically process images and This study proposes k-means clustering adaptive active contour model based...
Although Deep Learning (DL) models have been introduced in various fields as effective prediction tools, they often do not care about uncertainty. This can be a barrier to their adoption real-world applications. The current paper aims apply and evaluate Monte Carlo (MC) dropout, computationally efficient approach, investigate the reliability of several skip connection-based Convolutional Neural Network (CNN) while keeping high accuracy. To so, high-dimensional regression problem is...
This article presents detailed experimental results of the influencing factors flow rate, set pressure, and inlet pipe length a pressure relief valve. In order to analyze influence different rates on instability characteristics valve, multi-stage output pump is realized. terms modeling, we investigated theory concerning in system: 3% rule boundary. Data analyses typical stable, cycling, chatter conditions are conducted. The stable boundaries quarter-wave model drawn, which consistent with results.
Human chromosome classification is essential to the clinical diagnosis of cytogenetical diseases such as genetic disorders and cancer. This process, however, time-consuming requires specialist knowledge. Considerable efforts have been made automat process. Recently, methods based on Convolutional Neural Networks achieved state-of-the-art results task. Many studies used karyotype images in performance evaluation, few reported human microscopical images. paper proposes a novel method classify...
Abstract Trait‐based methods are key to understanding the biodiversity–productivity relationship (BPR) of macrophyte communities. Community‐weighted mean traits (i.e., community trait structure) have been proven more influence on productivity than species richness. However, underlying mechanism by which structure variation affects along an environmental gradient is still not well understood. A mesocosm experiment was used investigate how shapes water depth gradient. Three submerged (...