A5102747424- Advanced Neural Network Applications
- COVID-19 diagnosis using AI
- AI in cancer detection
- Domain Adaptation and Few-Shot Learning
- Brain Tumor Detection and Classification
- Radiomics and Machine Learning in Medical Imaging
- Traumatic Brain Injury and Neurovascular Disturbances
- Nonlinear Differential Equations Analysis
- Immune Response and Inflammation
- Face recognition and analysis
- Differential Equations and Numerical Methods
- Human Pose and Action Recognition
- dental development and anomalies
- Generative Adversarial Networks and Image Synthesis
- ECG Monitoring and Analysis
- Differential Equations and Boundary Problems
- Medical and Biological Sciences
- Optical measurement and interference techniques
- Advanced Image and Video Retrieval Techniques
- Oral microbiology and periodontitis research
- Visual Attention and Saliency Detection
- Cerebrovascular and Carotid Artery Diseases
- EEG and Brain-Computer Interfaces
- Infrared Target Detection Methodologies
- Advanced Vision and Imaging
University of Oulu
2021-2025
Shenzhen Second People's Hospital
2019-2024
Wuhan University
2022
Renmin Hospital of Wuhan University
2022
Hubei Zhongshan Hospital
2022
International Craniofacial Institute
2022
Hangzhou Dianzi University
2021-2022
Guangzhou University
2022
National University of Defense Technology
2020-2021
Sanda University
2021
Different from general face recognition, age-invariant recognition (AIFR) aims at matching faces with a big age gap. Previous discriminative methods usually focus on decomposing facial feature into age-related and components, which suffer the loss of identity information. In this article, we propose novel Multi-feature Fusion Decomposition (MFD) framework for learns more robust features reduces intra-class variants. Specifically, first sample multiple images different ages same as time...
Abstract Remote photoplethysmography (rPPG), which aims at measuring heart activities and physiological signals from facial video without any contact, has great potential in many applications (e.g., remote healthcare affective computing). Recent deep learning approaches focus on mining subtle rPPG clues using convolutional neural networks with limited spatio-temporal receptive fields, neglect the long-range perception interaction for modeling. In this paper, we propose two end-to-end...
Recently, there have been tremendous efforts in developing lightweight Deep Neural Networks (DNNs) with satisfactory accuracy, which can enable the ubiquitous deployment of DNNs edge devices. The core challenge compact and efficient lies how to balance competing goals achieving high accuracy efficiency. In this paper we propose two novel types convolutions, dubbed \emph{Pixel Difference Convolution (PDC) Binary PDC (Bi-PDC)} enjoy following benefits: capturing higher-order local differential...
Change captioning aims to describe the differences between two similar images using natural language, significantly aiding in understanding and monitoring changes. This challenging task requires a fine-grained of subtle changes while resisting disturbances like viewpoint shifts illumination variations. Existing methods often rely solely on global difference features lack comprehensive alignment linguistic visual information, leading overlooking details generating semantic hallucinated...
Programmed death 1 ligand (PD‑L1) is a negative co‑stimulatory molecule in immune responses. Previous reports have indicated that inflammatory cytokines can upregulate the expression of PD‑L1 tumor cells, which turn suppresses host Periodontitis characterized by persistent inflammation periodontium, initiated infection with oral bacteria and results damage to cells matrices periodontal connective tissues. In present study, function tissue destruction were examined. Periodontal ligament...
Accurately estimating the 3D pose and shape is an essential step towards understanding animal behavior, can potentially benefit many downstream applications, such as wildlife conservation. However, research in this area held back by lack of a comprehensive diverse dataset with high-quality annotations. In paper, we propose Animal3D, first for mammal estimation. Animal3D consists 3379 images collected from 40 species, annotations 26 key-points, importantly parameters SMAL [50] model. All were...
Binary neural networks (BNNs) constrain weights and activations to +1 or -1 with limited storage computational cost, which is hardware-friendly for portable devices. Recently, BNNs have achieved remarkable progress been adopted into various fields. However, the performance of sensitive activation distribution. The existing utilized Sign function predefined learned static thresholds binarize activations. This process limits representation capacity since different samples may adapt unequal...
Background Although it has been established that elevated blood pressure and its variability worsen outcomes in spontaneous intracerebral hemorrhage, antihypertensives use during the acute phase still lacks robust evidence. A pressure–lowering regimen using remifentanil dexmedetomidine might be a reasonable therapeutic option given their analgesic antisympathetic effects. The objective of this superiority trial was to validate efficacy safety strategy uses patients with hemorrhage. Methods...
Abstract As a distributed public ledger technology in peer‐to‐peer network, blockchain has been widely used variety of Internet interaction systems recent years, such as market supervision, property rights protection, and digital identity. However, environment, all historical transaction data are open transparent, adversaries may illegally access private data, which makes privacy attacks core issue that hinders the popularization application blockchain. Different privacy‐preserving...
Without developing dedicated countermeasures, facial biometric systems can be spoofed with printed photos, replay attacks, silicone masks, or even a 3D mask of targeted person. Thus, the threat presentation attacks needs to addressed strengthen security systems. Since 2D convolutional neural network (CNN) captures static features from video frames, camera motion might hinders performance modern CNNs for video-based attack detection (PAD). Inspired by egomotion theory, we introduce an...
Circular RNAs (circRNAs) are closely associated with the progression of oral squamous cell carcinoma (OSCC). circRNA_0001971 has been proved to accelerate OSCC development. Here, we aim identify new molecular mechanism hsa_circRNA_0001971 (circRNA_0001971) in OSCC.The levels circRNA_0001971, miR-186-5p, and fibronectin type III domain containing 3B (FNDC3B) tissues cells were verified by qRT-PCR or Western blotting. The interaction between FNDC3B was identified bioinformatics analysis,...
As is well-known, defects precisely affect the lives and functions of machines in which they occur, even cause potentially catastrophic casualties. Therefore, quality assessment before mounting an indispensable requirement for factories. Apart from recognition accuracy, current networks suffer excessive computing complexity, making it great difficulty to deploy manufacturing process. To address these issues, this paper introduces binary into area surface defect detection first time, reason...
Face recognition is one of the most active tasks in computer vision and has been widely used real world. With great advances made convolutional neural networks (CNN), lots face algorithms have achieved high accuracy on various datasets. However, existing based CNNs are vulnerable to noise. Noise corrupted image patterns could lead false activations, significantly decreasing noisy situations. To equip with built-in robustness noise different levels, we proposed a Median Pixel Difference...
The adaptive response of the mandible and temporomandibular joint (TMJ) to altered occlusion in juvenile patients is presently unclear. To address this question, we established a mouse model which all molars were extracted from maxillary right quadrant prepubertal, 3-week-old mice analyzed morphological, tissue, cellular, molecular changes condyle 3 weeks later. Unilateral loss led significant, robust, bilateral changes, primarily condylar morphology, including anteroposterior narrowing head...
This article proposes a novel module called middle spectrum grouped convolution (MSGC) for efficient deep convolutional neural networks (DCNNs) with the mechanism of convolution. It explores broad "middle spectrum" area between channel pruning and conventional Compared pruning, MSGC can retain most information from input feature maps due to group mechanism; compared convolution, benefits learnability, core constructing its topology, leading better division. The is unfolded along four...
Accurate segmentation of lung cancer in pathology slides is a critical step improving patient care. We proposed the ACDC@LungHP (Automatic Cancer Detection and Classification Whole-slide Lung Histopathology) challenge for evaluating different computer-aided diagnosis (CADs) methods on automatic cancer. The 2019 focused (pixel-wise detection) tissue whole slide imaging (WSI), using an annotated dataset 150 training images 50 test from 200 patients. This paper reviews this summarizes top 10...