检索结果-南通市图书馆

Gene regulatory network inference based on causal discovery integrating with graph neural network

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Quantitative Biology 2023年第4期11卷 434-450页

作者： Ke Feng Hongyang Jiang Chaoyi Yin Huiyan Sun School of artificial intelligence Jilin UniversityChangchunChina international center of Future Science Jilin UniversityChangchunChina Engineering research center of Knowledge-Driven Human-Machine intelligence Ministry of EducationChangchunChina

Gene regulatory network (GRN) inference from gene expression data is asignificant approach to understanding aspects of the biological system.Compared with generalized correlation-based methods, causality-inspiredones seem more rational to infer regulatory relationships. We proposeGRINCD, a novel GRN inference framework empowered by graph representationlearning and causal asymmetric learning, considering both linearand non-linear regulatory relationships. First, high-quality representation ofeach gene is generated using graph neural network. Then, we apply theadditive noise model to predict the causal regulation of each regulator-targetpair. Additionally, we design two channels and finally assemble them forrobust prediction. Through comprehensive comparisons of our frameworkwith state-of-the-art methods based on different principles on numerousdatasets of diverse types and scales, the experimental results show that ourframework achieves superior or comparable performance under variousevaluation metrics. Our work provides a new clue for constructing GRNs,and our proposed framework GRINCD also shows potential in identifyingkey factors affecting cancerdevelopment.

关键词： causal discovery ensemble learning gene regulatory network inference gene regulatory networks graph neural network key regulators of disease development

缺血性卒中后感染介导了低密度脂蛋白胆固醇与全因死亡风险之间的U形曲线关系

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science Bulletin 2023年第12期68卷 1327-1335,M0004页

作者：陈子墨谷鸿秋莫荆麟杨凯璇姜英玉杨昕王春娟许杰孟霞姜勇李昊刘丽萍王伊龙赵性泉李子孝王拥军 Beijing Tiantan Hospital Capital Medical UniversityBeijing 100071China China National Clinical research center for Neurological Diseases Beijing Tiantan HospitalCapital Medical UniversityBeijing 100071China National center for Healthcare Quality Management in Neurological Diseases Beijing Tiantan HospitalCapital Medical UniversityBeijing 100071China Advanced Innovation center for Human Brain Protection Capital Medical UniversityBeijing 100071China research Unit of artificial intelligence in Cerebrovascular Disease Chinese Academy of Medical SciencesBeijing 100071China center for Excellence in Brain Science and intelligence Technology Chinese Academy of SciencesShanghai 200031China

在缺血性卒中急性期,尚不清楚低水平的低密度脂蛋白胆固醇(LDL-C)的病理生理意义,本研究旨在评估LDL-C水平、卒中后感染和全因死亡间的关系.研究共纳入804,855名缺血性卒中患者.通过多元logistic回归模型分析LDL-C水平、感染和死亡之间... 详细信息

在缺血性卒中急性期,尚不清楚低水平的低密度脂蛋白胆固醇(LDL-C)的病理生理意义,本研究旨在评估LDL-C水平、卒中后感染和全因死亡间的关系.研究共纳入804,855名缺血性卒中患者.通过多元logistic回归模型分析LDL-C水平、感染和死亡之间的关系,采用基于反事实框架的中介分析,以阐明卒中后感染的中介效应,研究发现LDL-C与死亡风险之间呈U形曲线关系,死亡风险最低的LDL-C水平为2.67mmol/L.经多因素校正后,与LDL-C=2.50~2.99mmol/L相比,LDL-C<1.0mmol/L患者的死亡风险比值比(OR)为2.22(95%置信区间(CI):1.77~2.79),LDL-C5.0mmol/L患者的OR为1.22(95%CI:0.98~1.50).LDL-C与死亡之间关联的38.20%(95%CI:5.96~70.45,P=0.020)由感染介导:在逐步剔除存在心血管危险因素的患者后,LDL-C与全因死亡间的U形曲线关系及感染的中介效应与主要分析一致,而死亡风险最低时的LDL-C水平逐渐升高.在缺血性卒中急性期,LDL-C水平与全因死亡之间存在U形曲线关系,其被卒中后感染介导。

关键词：全因死亡死亡风险卒中后感染病理生理意义心血管危险因素多元logistic回归缺血性卒中低密度脂蛋白胆固醇

维普期刊数据库

Intelligent 3D garment system of the human body based on deep spiking neural network

在线全文

学校读者我要写书评

暂无评论

虚拟现实与智能硬件（中英文） 2024年第1期6卷 43-55页

作者： Minghua JIANG Zhangyuan TIAN Chenyu YU Yankang SHI Li LIU Tao PENG Xinrong HU Feng YU School of Computer and artificial intelligence Wuhan Textile UniversityWuhan 430200China Engineering research center of Hubei Province for Clothing Information Wuhan 430200China

Background Intelligent garments,a burgeoning class of wearable devices,have extensive applications in domains such as sports training and medical rehabilitation.Nonetheless,existing research in the smart wearables domain predominantly emphasizes sensor functionality and quantity,often skipping crucial aspects related to user experience and interaction.Methods To address this gap,this study introduces a novel real-time 3D interactive system based on intelligent garments.The system utilizes lightweight sensor modules to collect human motion data and introduces a dual-stream fusion network based on pulsed neural units to classify and recognize human movements,thereby achieving real-time interaction between users and sensors.Additionally,the system incorporates 3D human visualization functionality,which visualizes sensor data and recognizes human actions as 3D models in real time,providing accurate and comprehensive visual feedback to help users better understand and analyze the details and features of human motion.This system has significant potential for applications in motion detection,medical monitoring,virtual reality,and other fields.The accurate classification of human actions contributes to the development of personalized training plans and injury prevention strategies.Conclusions This study has substantial implications in the domains of intelligent garments,human motion monitoring,and digital twin visualization.The advancement of this system is expected to propel the progress of wearable technology and foster a deeper comprehension of human motion.

关键词： Intelligent garment system Internet of things Human action recognition Deep learning 3D visualization

维普期刊数据库

Abductive subconcept learning

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第2期66卷 114-126页

作者： Zhongyi HAN Le-Wen CAI Wang-Zhou DAI Yu-Xuan HUANG Benzheng WEI Wei WANG Yilong YIN School of Software Shandong University National Key Laboratory for Novel Software Technology Nanjing University Department of Computing Imperial College London center for medical artificial intelligence Shandong University of Traditional Chinese Medicine

Bridging neural network learning and symbolic reasoning is crucial for strong AI. Few pioneering studies have made some progress on logical reasoning tasks that require partitioned inputs of instances(e.g.,sequential data), from which a final concept is formed based on the complex(perhaps logical) relationships between them. However, they cannot apply to low-level cognitive tasks that require unpartitioned inputs(e.g., raw images), such as object recognition and text classification. In this paper, we propose abductive subconcept learning(ASL) to bridge neural network learning and symbolic reasoning on unsegmented image classification tasks. ASL uses deep learning and abductive logical reasoning to jointly learn subconcept perception and secondary reasoning. Specifically, it first employs meta-interpretive learning(MIL) to induce first-order logical hypotheses capturing the relationships between the high-level subconcepts that account for the target concept. Then, it uses the groundings of the logical hypotheses as labels to train a deep learning model for identifying the subconcepts from unpartitioned data. ASL jointly trains the deep learning model and learns the MIL theory by minimizing the inconsistency between their grounded outputs. Experimental results show that ASL successfully integrates machine learning and logical reasoning with accurate and interpretable results in several object recognition tasks.

关键词： abductive learning subconcept learning logical reasoning subconcept set selection meta-interpretive learning

Risk prediction platform for pancreatic fistula after pancreatoduodenectomy using artificial intelligence

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

World Journal of Gastroenterology 2020年第30期26卷 4453-4464页

作者： In Woong Han Kyeongwon Cho Youngju Ryu Sang Hyun Shin Jin Seok Heo Dong Wook Choi Myung Jin Chung Oh Chul Kwon Baek Hwan Cho Department of Surgery Samsung Medical CenterSungkyunkwan University School of MedicineSeoul 06351South Korea medical artificial intelligence research center Department of Medical Device Management and ResearchSAIHSTSamsung Medical CenterSungkyunkwan University School of MedicineSeoul 06351South Korea Department of Radiology Samsung Medical CenterSungkyunkwan University School of MedicineSeoul 06351South Korea artificial intelligence research center Medical DataBase IncorporatedSeoul 06048South Korea

BACKGROUND Despite advancements in operative technique and improvements in postoperative managements,postoperative pancreatic fistula(POPF)is a life-threatening complication following pancreatoduodenectomy(PD).There are some reports to predict POPF preoperatively or intraoperatively,but the accuracy of those is questionable.artificial intelligence(AI)technology is being actively used in the medical field,but few studies have reported applying it to outcomes after PD.AIM To develop a risk prediction platform for POPF using an AI model.METHODS medical records were reviewed from 1769 patients at Samsung medical center who underwent PD from 2007 to 2016.A total of 38 variables were inserted into AI-driven algorithms.The algorithms tested to make the risk prediction platform were random forest(RF)and a neural network(NN)with or without recursive feature elimination(RFE).The median imputation method was used for missing values.The area under the curve(AUC)was calculated to examine the discriminative power of algorithm for POPF prediction.RESULTS The number of POPFs was 221(12.5%)according to the International Study Group of Pancreatic Fistula definition 2016.After median imputation,AUCs using 38 variables were 0.68±0.02 with RF and 0.71±0.02 with NN.The maximal AUC using NN with RFE was 0.74.Sixteen risk factors for POPF were identified by AI algorithm:Pancreatic duct diameter,body mass index,preoperative serum albumin,lipase level,amount of intraoperative fluid infusion,age,platelet count,extrapancreatic location of tumor,combined venous resection,co-existing pancreatitis,neoadjuvant radiotherapy,American Society of Anesthesiologists’score,sex,soft texture of the pancreas,underlying heart disease,and preoperative endoscopic biliary decompression.We developed a web-based POPF prediction platform,and this application is freely available at http://popfrisk.***.org.CONCLUSION This study is the first to predict POPF with multiple risk factors using AI.This platform is reliable(AUC 0.74),so it could be used to select pat

关键词： Postoperative pancreatic fistula Pancreatoduodenectomy Neural networks Recursive feature elimination

Deep learning-based semantic segmentation of human features in bath scrubbing robots

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Biomimetic intelligence & Robotics 2024年第1期4卷 70-79页

作者： Chao Zhuang Tianyi Ma Bokai Xuan Cheng Chang Baichuan An Minghuan Yin Hao Sun School of artificial intelligence and Data Science Hebei University of TechnologyTianjin 300401China Engineering research center of Intelligent Rehabilitation Device and Detection Technology Ministry of EducationTianjin 300401China

With the rise in the aging population,an increase in the number of semidisabled elderly individuals has been noted,leading to notable challenges in medical and healthcare,exacerbated by a shortage of nursing staff.This study aims to enhance the human feature recognition capabilities of bath scrubbing robots operating in a water fog environment.The investigation focuses on semantic segmentation of human features using deep learning methodologies.Initially,3D point cloud data of human bodies with varying sizes are gathered through light detection and ranging to establish human models.Subsequently,a hybrid filtering algorithm was employed to address the impact of the water fog environment on the modeling and extraction of human regions.Finally,the network is refined by integrating the spatial feature extraction module and the channel attention module based on PointNet.The results indicate that the algorithm adeptly identifies feature information for 3D human models of diverse body sizes,achieving an overall accuracy of 95.7%.This represents a 4.5%improvement compared with the PointNet network and a 2.5%enhancement over mean intersection over union.In conclusion,this study substantially augments the human feature segmentation capabilities,facilitating effective collaboration with bath scrubbing robots for caregiving tasks,thereby possessing significant engineering application value.

关键词： 3D point cloud Human model LiDAR Semantic segmentation

Detailed phenotype of RNF213 p.R4810K variant identified by the Chinese patients with acute ischaemic stroke or transient ischaemic attack

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Stroke & Vascular Neurology 2023年第6期8卷 503-510页

作者： Hongyu Zhou Jing Jing Yuehua Pu Wei Li Xia Meng Anxin Wang Yingting Zuo Zhe Xu Qin Xu Yue Suo Hao Li Yongjun Wang Department of Neurology Beijing Tiantan HospitalCapital Medical UniversityBeijingChina China National Clinical research center for Neurological Diseases BeijingChina Tiantan Neuroimaging center of Excellence Beijing Tiantan HospitalCapital Medical UniversityBeijingChina Monogenic Disease Diagnosis center for Neurological Disorders and Precision Medicine research center for Neurological Disorders Beijing Tiantan HospitalCapital Medical UniversityBeijingChina Advanced Innovation center for Human Brain Protection Capital Medical UniversityBeijingChina center of excellence for Omics research(CORe) Beijing Tiantan HospitalCapital Medical UniversityBeijingChina research Unit of artificial intelligence in Cerebrovascular Disease Chinese Academy of Medical SciencesBeijingChina

Background and purpose The ring finger protein 213 gene(RNF213)p.R4810K variant increased the risk of acute ischaemic stroke(AIS)attributable to intracranial arterial stenosis(ICAS)in the Japanese and Korean populations.In this study,we aimed to examine the prevalence of the RNF213 p.R4810K variant in Chinese patients with AIS or transient ischaemic attack and identify the phenotype of the carriers.Methods We analysed data from the Third China National Stroke Registry.All included participants were divided into two groups by carrier status of the p.R4810K variant.The aetiological classification was conducted according to the Trial of Org 10172 in Acute Stroke Treatment(TOAST)criteria.The presence of ICAS and extracranial arterial stenosis(ECAS)was defined as 50%-99%stenosis or occlusion of any intracranial and extracranial artery.Logistic regression models and Cox regression models were used to evaluate the association of the p.R4810K variant with TOAST classification,stenosis phenotypes and clinical outcomes.Results A total of 10381 patients were enrolled,among which 56(0.5%)had the heterozygote GA genotype for p.R4810K.The variant carriers were younger(p=0.01),and more likely to suffer from peripheral vascular disease(p=0.04).The p.R4810K variant was associated with large-artery atherosclerosis(LAA)(adjusted OR=1.94,95%CI 1.13 to 3.33),anterior circulation stenosis(adjusted OR=2.12,95%CI 1.23 to 3.65)and ECAS(adjusted OR=2.29,95%CI 1.16 to 4.51).Nevertheless,the p.R4810K variant was not associated with recurrence,poor functional outcome and mortality at 3 months and 1 year.Conclusions The RNF213 p.R4810K variant was associated with LAA,anterior circulation stenosis and ECAS in Chinese patients.Given the low carrying rate and only 1-year follow-up information,caution should be taken to interpret our findings in no statistically significant association between the p.R4810K variant and stroke prognosis in Chinese patients.

关键词： patients transient younger

Collective and Individual Assessment of the Risk of Death from COVID-19 for the Elderly,2020–2022

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China CDC weekly 2023年第18期5卷 407-412,I0001,I0002页

作者： Chaobao Zhang Hongzhi Wang Zilu Wen Zhijun Bao Xiangqi Li Shanghai Key Laboratory of Clinical Geriatric Medicine Department of Geriatric MedicineHuadong HospitalShanghai Medical CollegeFudan UniversityShanghaiChina Shanghai Key Laboratory of Magnetic Resonance Research Center for Artificial Intelligence in Medical ImagingEast China Normal UniversityShanghaiChina Department of Scientific research Shanghai Public Health Clinical CenterShanghaiChina Department of Endocrinology and Metabolism Gongli HospitalNaval Medical UniversityShanghaiChina

Introduction:Coronavirus disease 2019(COVID-19)has had profound disruptions worldwide.For a population or individual,it is critical to assess the risk of death for making preventative decisions.Methods:In this study,clinical data from approximately 100 million cases were statistically analyzed.A software and an online assessment tool were developed in Python to evaluate the risk of mortality.Results:Our analysis revealed that 76.51%of COVID-19-related fatalities occurred among individuals aged over 65 years,with frailty-associated deaths accounting for more than 80%of these cases.Furthermore,over 80%of the reported deaths involved unvaccinated individuals.A notable overlap was observed between aging and frailty-associated deaths,both of which were connected to underlying health conditions.For those with at least two comorbidities,the proportion of frailty and the proportion of COVID-19-related death were both close to 75 percent.Subsequently,we established a formula to calculate the number of deaths,which was validated using data from twenty countries and regions.Using this formula,we developed and verified an intelligent software designed to predict the death risk for a given population.To facilitate rapid risk screening on an individual level,we also introduced a six-question online assessment tool.Conclusions:This study examined the impact of underlying diseases,frailty,age,and vaccination history on COVID-19-related mortality,resulting in a sophisticated software and a user-friendly online scale to assess mortality risk.These tools offer valuable assistance in informed decision-making.

关键词： Python deaths software

Knowledge Graph Enhanced Transformers for Diagnosis Generation of Chinese Medicine

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Integrative Medicine 2024年第3期30卷 267-276页

作者： WANG Xin-yu YANG Tao GAO Xiao-yuan HU Kong-fa School of artificial intelligence and Information Technology Nanjing University of Chinese MedicineNanjing210023China School of Information Management Nanjing UniversityNanjing210023China Jiangsu Collaborative Innovation center of Traditional Chinese Medicine in Prevention and Treatment of Tumor Nanjing210023China Jiangsu Province Engineering research center of TCM intelligence Health Service Nanjing210023China

Chinese medicine(CM)diagnosis intellectualization is one of the hotspots in the research of CM modernization.The traditional CM intelligent diagnosis models transform the CM diagnosis issues into classification issues,however,it is difficult to solve the problems such as excessive or similar categories.With the development of natural language processing techniques,text generation technique has become increasingly mature.In this study,we aimed to establish the CM diagnosis generation model by transforming the CM diagnosis issues into text generation issues.The semantic context characteristic learning capacity was enhanced referring to Bidirectional Long Short-Term Memory(BILSTM)with Transformer as the backbone network.Meanwhile,the CM diagnosis generation model Knowledge Graph Enhanced Transformer(KGET)was established by introducing the knowledge in medical field to enhance the inferential capability.The KGET model was established based on 566 CM case texts,and was compared with the classic text generation models including Long Short-Term Memory sequence-to-sequence(LSTM-seq2seq),Bidirectional and Auto-Regression Transformer(BART),and Chinese Pre-trained Unbalanced Transformer(CPT),so as to analyze the model manifestations.Finally,the ablation experiments were performed to explore the influence of the optimized part on the KGET model.The results of Bilingual Evaluation Understudy(BLEU),Recall-Oriented Understudy for Gisting Evaluation 1(ROUGE1),ROUGE2 and Edit distance of KGET model were 45.85,73.93,54.59 and 7.12,respectively in this study.Compared with LSTM-seq2seq,BART and CPT models,the KGET model was higher in BLEU,ROUGE1 and ROUGE2 by 6.00–17.09,1.65–9.39 and 0.51–17.62,respectively,and lower in Edit distance by 0.47–3.21.The ablation experiment results revealed that introduction of BILSTM model and prior knowledge could significantly increase the model performance.Additionally,the manual assessment indicated that the CM diagnosis results of the KGET model used in this study were highly consistent with the p

关键词： Chinese medicine diagnosis knowledge graph enhanced transformer text generation

维普期刊数据库