A5100326679- Gene Regulatory Network Analysis
- Bioinformatics and Genomic Networks
- Toxic Organic Pollutants Impact
- Ecosystem dynamics and resilience
- Single-cell and spatial transcriptomics
- COVID-19 epidemiological studies
- Air Quality and Health Impacts
- Mental Health Research Topics
- Influenza Virus Research Studies
- Metabolomics and Mass Spectrometry Studies
- Topic Modeling
- Microplastics and Plastic Pollution
- Recycling and Waste Management Techniques
- Renal Diseases and Glomerulopathies
- Sleep and related disorders
- Natural Language Processing Techniques
- Neural Networks and Applications
- Data-Driven Disease Surveillance
- Adsorption and biosorption for pollutant removal
- Caching and Content Delivery
- Advanced Memory and Neural Computing
- Infections and bacterial resistance
- Complex Systems and Time Series Analysis
- Bacterial Identification and Susceptibility Testing
- Mycobacterium research and diagnosis
Central South University
2016-2025
South China University of Technology
2015-2025
University of Illinois Chicago
2022-2025
Xiangya Hospital Central South University
2021-2025
Fudan University
2008-2025
Shanghai Fudan Microelectronics (China)
2025
Shanghai Zhangjiang Laboratory
2025
Tianjin University
2024-2025
Texas A&M University
2021-2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2022-2024
Abstract Background Developing effective strategies for signaling the pre-disease state of complex diseases, a with high susceptibility before disease onset or deterioration, is urgently needed because such usually followed by catastrophic transition into worse stage disease. However, it challenging task to identify tipping point in clinics, where only one single sample available and thus results failure most statistic approaches. Methods In this study, we presented single-sample-based...
Multimode-fused sensing in the somatosensory system helps people obtain comprehensive object properties and make accurate judgments. However, building such multisensory systems with conventional metal-oxide-semiconductor technology presents serious device integration circuit complexity challenges. Here, a multimode-fused spiking neuron (MFSN) compact structure to achieve human-like perception is reported. The MFSN heterogeneously integrates pressure sensor process NbOx -based memristor sense...
Natural language processing (NLP) and machine learning were used to predict suicidal ideation heightened psychiatric symptoms among adults recently discharged from inpatient or emergency room settings in Madrid, Spain. Participants responded structured mental physical health instruments at multiple follow-up points. Outcome variables of interest (GHQ-12). Predictor included items (e.g., relating sleep well-being) responses one unstructured question, “how do you feel today?” We compared...
Abstract Identifying early-warning signals of a critical transition for complex system is difficult, especially when the target constantly perturbed by big noise, which makes traditional methods fail due to strong fluctuations observed data. In this work, we show that not state-transition but probability distribution-transition noise sufficiently small, which, however, ubiquitous case in real systems. We present model-free computational method detect warning before such transitions. The key...
We develop an auto-reservoir computing framework, Auto-Reservoir Neural Network (ARNN), to efficiently and accurately make multi-step-ahead predictions based on a short-term high-dimensional time series. Different from traditional reservoir whose is external dynamical system irrelevant the target system, ARNN directly transforms observed dynamics as its reservoir, which maps high-dimensional/spatial data future temporal values of variable our spatiotemporal information (STI) transformation....
Background Typically, algorithms to classify phenotypes using electronic medical record (EMR) data were developed perform well in a specific patient population. There is increasing interest analyses which can allow study of outcome across different diseases. Such the EMR would require an algorithm that be applied populations. Our objectives were: (1) develop enable coronary artery disease (CAD) diverse populations; (2) impact adding narrative extracted natural language processing (NLP)...
Abstract Continuous glucose monitoring (CGM) technology has grown rapidly to track real-time blood levels and trends with improved sensor accuracy. The ease of use wide availability CGM would facilitate safe effective decision making for diabetes management. Here, we developed an attention-based deep learning model, CGMformer, pretrained on a well-controlled diverse corpus data represent individual's intrinsic metabolic state enable clinical applications. During pretraining, CGMformer...
Abstract Acquired drug resistance is the major reason why patients fail to respond cancer therapies. It a challenging task determine tipping point of endocrine and detect associated molecules. Derived from new systems biology theory, dynamic network biomarker (DNB) method designed quantitatively identify drastic system transition can theoretically DNB genes that play key roles in acquiring resistance. We analyzed time-course mRNA sequence data generated tamoxifen-treated estrogen receptor...
Objective Cohort selection is challenging for large-scale electronic health record (EHR) analyses, as International Classification of Diseases 9th edition (ICD-9) diagnostic codes are notoriously unreliable disease predictors. Our objective was to develop, evaluate, and validate an automated algorithm determining Autism Spectrum Disorder (ASD) patient cohort from EHR. We demonstrate its utility via the largest investigation date co-occurrence patterns medical comorbidities in ASD. Methods...
Identifying the critical state or pre-transition just before occurrence of a phase transition is challenging task, because system may show little apparent change this during gradual parameter variations. Such dynamics generally composed three stages, i.e. before-transition state, and after-transition which can be considered as different Markov processes.By exploring rich dynamical information provided by high-throughput data, we present novel computational method, hidden model (HMM) based...
The time evolution or dynamic change of many biological systems during disease progression is not always smooth but occasionally abrupt, that is, there a tipping point such process at which the system state shifts from normal to state. It challenging predict with measured omics data, in particular when only single sample available.In this study, we developed novel approach, i.e. single-sample landscape entropy (SLE) method, identify one data. Specifically, by evaluating disorder network...
Complex diseases do not always follow gradual progressions. Instead, they may experience sudden shifts known as critical states or tipping points, where a marked qualitative change occurs. Detecting such pivotal transition pre-deterioration state holds paramount importance due to its association with severe disease deterioration. Nevertheless, the task of pinpointing for complex remains an obstacle, especially in scenarios involving high-dimensional data limited samples, conventional...
Key Points A multiancestry proteome-wide Mendelian randomization analysis was conducted for IgA nephropathy. The findings from the study would help prioritize new drug targets and drug-repurposing opportunities. Background therapeutic options nephropathy are rapidly evolving, but early diagnosis targeted treatment remain challenging. We aimed to identify circulating plasma proteins associated with by studies across multiple ancestry populations. Methods In this study, we applied...
Abstract The seasonal outbreaks of influenza infection cause globally respiratory illness, or even death in all age groups. Given early‐warning signals preceding the outbreak, timely intervention such as vaccination and isolation management effectively decrease morbidity. However, it is usually a difficult task to achieve real‐time prediction outbreak due its complexity intertwining both biological systems social systems. By exploring rich dynamical high‐dimensional information, our dynamic...
The progression of complex diseases, such as diabetes and cancer, is generally a nonlinear process with three stages, i.e., normal state, pre-disease disease where the state critical or tipping point immediately preceding state. Traditional biomarkers aim to identify by exploiting information differential expressions for observed molecules, but may fail detect because there are little significant differences between states. Thus, it challenging signal which actually implies prediction. In...