- TGF-β signaling in diseases
- Advanced Topology and Set Theory
- Cerebrovascular and genetic disorders
- Advanced Banach Space Theory
- Speech Recognition and Synthesis
- Topic Modeling
- Algebraic Geometry and Number Theory
- Higher Education and Teaching Methods
- Advanced Algebra and Geometry
- Music and Audio Processing
- Biochemical effects in animals
- Migraine and Headache Studies
- Virtual Reality Applications and Impacts
- Biocrusts and Microbial Ecology
- Neural Networks and Applications
- Education and Learning Interventions
- AI in cancer detection
- Natural Language Processing Techniques
- Skin Protection and Aging
- Drug Transport and Resistance Mechanisms
- Homotopy and Cohomology in Algebraic Topology
- Wikis in Education and Collaboration
- Advanced Malware Detection Techniques
- Flavonoids in Medical Research
- Computational Drug Discovery Methods
Shaoxing University
2025
Zhejiang University
2022-2025
Center for Life Sciences
2025
Johns Hopkins University
2021-2024
Syracuse University
2024
Virginia Tech
2024
University of California, Berkeley
2024
Dallas Independent School District
2024
Jiangsu University
2023-2024
Taiyuan University of Technology
2024
Migraine is a disabling neurological disorder with the pathophysiology yet to be understood. The microstructural alteration in brain white matter (WM) has been suggested related migraine recent studies, but these evidence are observational essentially and cannot infer causal relationship. present study aims reveal relationship between WM using genetic data Mendelian randomization (MR).
Artemisinin has an endoperoxide bridge structure, which can be cleaved by ferrous ions to generate various carbonyl radicals in oxygen-independent manner, highlighting its potential for treating hypoxic tumors. In our study, we fabricated Tween 80 micelles loaded with Fe3O4 nanoparticles and artemisinin cancer therapy. The synthesized drug-loaded have particle sizes of about 5 nm nm, respectively, both exhibiting excellent dispersibility stability. After uptake MCF-7 cells, release Fe2+ ART...
This research aims to explore the integration method of deep learning and large language models in speech recognition improve system’s accuracy ability handle complex contexts. Deep neural network (DNN), convolutional (CNN), long short-term memory (LSTM) Transformer-based model are used build an integrated acoustic framework. Experiments on TIMIT, LibriSpeech Common Voice datasets show that ensemble shows significant improvements both word error rate (WER) real-time factor (RTF)...
Apigenin (APG) is a well-known dietary flavonoid with multiple bioactivities, but its poor aqueous solubility may result in low oral bioavailability and thus compromised therapeutic effects. In the present study, APG was complexed oxymatrine (OMT), natural quinolizidine alkaloid, for enhanced anti-inflammatory activity, related mechanisms interaction of OMT were investigated. Fourier transform-infrared spectroscopy, fluorescence Raman proton nuclear magnetic resonance spectroscopy...
Ferroptosis therapy efficacy of cancers suffers from relatively low concentrations intracellular free iron ions due to the efficient regulation through storage in ferritin and efflux via ferroportin (FPN). In this study, a ferritin/ferroportin-hijacking nanoplatform (Fe3O4-ART@MM-Hep) containing artemisinin (ART) hepcidin (Hep) is fabricated boost induce reactive oxygen species (ROS) storm. Once tumor site reached, targeted binds FPN triggers internalization degradation FPN, blocking ions....
Corrected by: Comparisons of Scutellaria baicalensis, lateriflora and racemosa: Genome Size, Antioxidant Potential PhytochemistryPlanta Med 2008; 74(04): 482-482DOI: 10.1055/s-2008-1074507
Abstract Background UV radiation (UV) exposure risks skin damage and cancer due to DNA oxidative stress. Synthetic chemical sunscreens that protect against can have health environmental concerns. This study explores phycocyanin (PC), a marine algae-derived natural photoprotective compound, its crosslinked nanoparticles (PCNP) as safe effective adjuvants for sunscreen systems. Methods PCNP was synthesized via genipin-crosslinking. PC biocompatibility were assessed on mouse embryonic...
This research aims to explore the integration method of deep learning and large language models in speech recognition improve system’s accuracy ability handle complex contexts. Deep neural network (DNN), convolutional (CNN), long short-term memory (LSTM) Transformer-based model are used build an integrated acoustic framework. Experiments on TIMIT, LibriSpeech Common Voice datasets show that ensemble shows significant improvements both word error rate (WER) real-time factor (RTF)...
Leukemia is a fatal disease that requires the counting of White Blood Cells (WBCs) in bone marrow for diagnosis. However, blood contains many types leukocytes, some which have subtle differences. To address this issue, we propose WBC-GLAformer model, comprises three parts: Low-level Feature Extractor (LFE), Global-Local Attention based Encoder (GLAE), and Discrimination Part Select (DPS). The LFE uses convolutional neural network (CNN) to tokenize patches from extracted low-level features....
This research aims to explore the integration method of deep learning and large language models in speech recognition improve system’s accuracy ability handle complex contexts. Deep neural network (DNN), convolutional (CNN), long short-term memory (LSTM) Transformer-based model are used build an integrated acoustic framework. Experiments on TIMIT, LibriSpeech Common Voice datasets show that ensemble shows significant improvements both word error rate (WER) real-time factor (RTF)...
Reactive oxygen species (ROS) have been implicated in the pathogenesis of many clinical disorders such as adult respiratory distress syndrome, ischemia-reperfusion injury, atherosclerosis, neurodegenerative diseases, and cancer.Genetically engineered animal models used a tool for understanding function various antioxidant enzymes cellular defense mechanisms against types oxidant tissue injury.Transgenic mice overexpressing three isoforms superoxide dismutase, catalase, glutathione peroxidase...
Modern technology advances airplane seat design with better ergonomics and new HCI (human-computer interaction). However, airline companies are not motivated to replace the system due cost consideration. Hence, a series of re-optimized in should be carried out by designers. This paper describes novel intelligent seat’s system, which is designed used for airplanes or similar conditions. consists redesigned compared original systems. The mainly parts aesthetics visual modeling people receive...
This paper introduces the concept of an intelligent vase that combines cultivation plants with psychotherapy for Alzheimer’s adjuvant therapy (from now on, referred to as AD) patients. RECALL (The name which defined by this paper), includes hardware and software applications, well a unique hand gesture interaction method, is able collect patients’ gestures play photos, remind patients plant’s growth status. To evaluate our conceptual system, we created preliminary prototype. We conducted...
Modern virtual personal assistants provide a convenient interface for completing daily tasks via voice commands. An important consideration these is the ability to recover from automatic speech recognition (ASR) and natural language understanding (NLU) errors. In this paper, we focus on learning robust dialog policies To end, develop user simulator which interacts with assistant through commands in realistic scenarios noisy audio, use it learn deep reinforcement learning. We show that...
Understanding human language often necessitates understanding entities and their place in a taxonomy of knowledge—their types.Previous methods to learn entity types rely on training classifiers datasets with coarse, noisy, incomplete labels. We introduce method instill fine-grained type knowledge models text-to-text pre-training type-centric questions leveraging base documents graphs.We create the WikiWiki dataset: passages from 10M Wikipedia articles linked Wikidata graph 41K types.Models...