泛研网

公众版 查看说明
全球科研项目数据库2026版

神经系统疾病专病队列研究

项目来源

国家重点研发计划(NKRD)

项目主持人

笪宇威

项目受资助机构

首都医科大学宣武医院

立项年度

2017

立项时间

未公开

项目编号

2017YFC0907700

项目级别

国家级

研究期限

未知 / 未知

受资助金额

1417.00万元

学科

精准医学研究

学科代码

未公开

基金类别

精准医学研究重点专项

关键词

神经系统疾病 ; 队列研究 ; 表型数据库 ; 生物样本库 ; 精准医疗 ; neurological diseases ; cohort study ; clinical phenotypes database ; biological samples database ; precision medicine

参与者

朵建英

参与机构AI

空军军医大学;中山大学;首都医科大学;广州中医药大学;河北医科大学;西安医学院

项目标书摘要：癫痫、阿尔茨海默病、帕金森病、多发性硬化/视神经脊髓炎谱系疾病和重症肌无力是严重危害我国居民健康的重大神经系统疾病，其发病机制、病因与危险因素、精确诊断和分型、个体化防诊治策略等尚不明确，缺乏统一标准及流程的大型神经疾病专病队列。本研究整合现有6种疾病临床表型数据库资源，建立神经系统疾病专病社区队列及大样本多中心临床队列，建立数据库及生物样本库共享平台，最终形成规模达到3.4 万社区人群队列及2 万例患者的临床队列，并长期随访，明确各疾病表型特征，认识疾病演变规律与预后，支撑相关疾病分子分型与精准化诊治方案研究。现阶段工作已完成社区及临床队列的建立，其中社区队列及癫痫、重症肌无力、脱髓鞘疾病队列均超额完成任务，各疾病队列建立统一标准和规范的电子数据库及完善、规范化的样本库，创建了新型、高效的随访体系，严格实施组织一体化，各课题间共同协作、定期研讨，中期任务目标达成。下一阶段工作将围绕已有数据的补充完善、数据库和样本库的联动、随访工作来开展，并开展第三方机构监察工作；在上述工作基础上，针对癫痫、阿尔兹海默病/帕金森病、多发性硬化/视神经脊髓炎谱疾病、重症肌无力的精准分型、预测因素、精准治疗等展开研究，为后续研究提供详实疾病信息及生物样本，研究成果用于诊断、治疗流程及指南的制定工作。

Application Abstract: It still remains unclear about the pathogenesis,risk factors,accurate classification and personalized treatment strategies of epilepsy,Alzheimer’s disease,Parkinson’s disease and myasthenia gravis(MG).The nationwide cohort studies of Epilepsy(EP),Alzheimer’s Disease(AD),Parkinson Disease(PD),Myasthenia Gravis(MG),Multiple Sclerosis(MS)and Optic Neuromyelitis Spectrum Disease are lack in China.This research,integrating those diseases clinical database resources in existence from 6 centers,to establish neurology disease’community cohort and large sample multiple centers clinical cohort.Ultimately a community cohort of 3,4000 patients and clinical cohort of 20,000 patients will be established,with long-term follow-up.The database of detailed clinical phenotypes and biological samples can be shared for setting up accurate diagnosis,classification and therapy and follow-up system.At this stage,the establishment of community and clinical cohort has been completed.The community cohort and the EP,MG,and demyelinating disease cohort have all completed the tasks.Each disease queue has established a unified standard and standardized electronic database and a complete and standardized sample library.A new and efficient follow-up system has been established,and organizational integration has been strictly implemented.Collaboration and regular discussions have been held among various topics,and the medium-term tasks have been achieved.The next phase of work will focus on the complementation of existing data,linkage of database and sample libraries,follow-up work,and third-party agency monitoring;based on the above work,precise classification,predictive factors,precision treatment of these disease will provide detailed disease information and biological samples for follow-up research.

项目受资助省

北京市

  • 排序方式：
  • 1
  • /

  • 1.A Deep Learning Framework to Predict Rating for Cold Start Item Using Item Metadata

    • 关键词：
    • User profile;Deep learning;Text processing;Recommender systems;Auto encoders;Cold start;Feature extraction techniques;Performance of recommender systems;profile Learner;State-of-the-art approach;Term frequencyinverse document frequency (TF-IDF);Word2vec
    • Anwar, Fahad;Iltaf, Naima;Afzal, Hammad;Abbas, Haider
    • 《28th IEEE International Conference on Enabling Technologies: Infrastructure for Collaborative Enterprises, WETICE 2019》
    • 2019年
    • June 12, 2019 - June 14, 2019
    • Capri, Italy
    • 会议

    Recommender systems improve browsing experience of users for large amount of items by assisting selection and classification of items utilizing item metadata. The performance of recommender system usually deteriorates when implicit data is used with limited user interaction history also regarded as cold start (CS) problem. This paper proposes a model to address cold start problem using content based technique where user or item metadata is used to break this ice barrier. The proposed method utilizes the feature extraction techniques (such as term frequencyInverse document frequency(TF-IDF)) and word embedding technique (Word2Vec). These content features are then used to predict the ratings for CS items by constructing user profiles using stacked auto-encoder. Experiments performed on largest real world dataset provided by Movielens 20M shows that proposed model outperforms the state-of-the-art approaches in CS item scenario. © 2019 IEEE.

    ...
  • 排序方式：
  • 1
  • /