III: Small: Mining Heterogeneous Network Built from Multiple Data Sources to Reduce Opioid Overdose Risks

III：小型：挖掘由多个数据源构建的异构网络以减少阿片类药物过量风险

• 批准号: 1951504
• 负责人: Yanfang Ye
• 金额: $ 48.95万
• 依托单位: Case Western Reserve University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-05 至 2022-02-28
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1951504&HistoricalAwards=false
• 关键词: III; Small; Mining; Heterogeneous; Network

As opioid overdose deaths have continued to increase over the past decade across the country, it is critical to understand the drugs involved in those deaths and the potential role of polypharmacy (i.e., the concurrent use of multiple medications) in opioid overdose deaths. However, due to the formidable complexity of drug-drug interactions (DDIs) arising from polypharmacy, it is challenging if not impossible to count them all manually. Therefore, there is an urgent need for developing novel computational methodologies and models for early detection of risky DDI patterns when opioids are combined with other drugs (e.g., sedatives, muscle relaxants, anti-anxieties). Since relying on a single data source for biomedical knowledge discovery often results in unsatisfactory performance, the goal of this project is to design and develop a novel and integrated framework (algorithms, models, and techniques) to construct a heterogeneous network built from multiple data sources and extract useful information from the constructed network to reduce the risk of opioid overdoses resulting from polypharmacy. The key components of the planned research are three-folds. First, the research team will construct a heterogeneous network from multiple data sources for abstract representation. Second, the team will develop scalable techniques for large-scale and dynamic heterogeneous network representation learning. Third, the team will design a novel deep learning framework with interpretability enhancement for early detection of risky DDI patterns when opioids are combined with other medications. The broad impacts of this work include benefits to the general public by addressing one of the most challenging issues facing U.S. public health today (i.e., overdose death prevention). The planned research in this project is also beneficial to the intelligent information management domain where multiple data sources are involved. The project integrates interdisciplinary research with education through curriculum development, the participation of underrepresented groups, and student mentoring activities.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

由于阿片类药物过量死亡在过去十年中在全国范围内持续增加，了解这些死亡涉及的药物以及多药联用(即同时使用多种药物)在阿片类药物过量死亡中的潜在作用至关重要。然而，由于多药联用引起的药物-药物相互作用(DDiS)的巨大复杂性，手动统计它们即使不是不可能的话也是具有挑战性的。因此，迫切需要开发新的计算方法和模型，以便在阿片类药物与其他药物(如镇静剂、肌肉松弛药、抗焦虑药)联合使用时，早期发现危险的DDI模式。由于生物医学知识发现依赖单一的数据源往往会导致性能不佳，本项目的目标是设计和开发一个新颖的集成框架(算法、模型和技术)，以构建一个由多个数据源构建的异质网络，并从所构建的网络中提取有用信息，以降低因多药而导致的阿片类药物过量的风险。计划中的研究的关键组成部分是三个方面。首先，研究团队将从多个数据源构建一个异构网络，用于抽象表示。其次，该团队将开发可扩展的技术，用于大规模和动态的异质网络表示学习。第三，该团队将设计一个新的深度学习框架，提高可解释性，以便在阿片类药物与其他药物联合使用时早期检测危险的DDI模式。这项工作的广泛影响包括解决当今美国公共卫生面临的最具挑战性的问题之一(即预防过量死亡)，从而使普通公众受益。本项目计划的研究对于涉及多个数据源的智能信息管理领域也是有益的。该项目通过课程开发、代表不足的群体的参与和学生指导活动，将跨学科研究与教育相结合。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Network Schema Preserved Heterogeneous Information Network Embedding

网络模式保留异构信息网络嵌入

• 发表时间: 2020
• 期刊: 29th International Joint Conference on Artificial Intelligence (IJCAI
• 作者: Zhao, Jianan;Wang, Xiao;Shi, Chuan;Liu, Zekuan;Ye, Yanfang.
• 通讯作者: Ye, Yanfang.

Joint Demosaicing and Super-Resolution (JDSR): Network Design and Perceptual Optimization

• DOI: 10.1109/tci.2020.2999819
• 发表时间: 2019-11
• 期刊: IEEE Transactions on Computational Imaging
• 影响因子: 5.4
• 作者: Xuan Xu;Yanfang Ye;Xin Li

The Impact of Rural Classification Systems on a Comparison of Risky Drug-Related Behaviors in Kentucky and Ohio Counties.

农村分类系统对肯塔基州和俄亥俄州危险药物相关行为比较的影响。

• DOI: 10.1080/10911359.2018.1516591
• 发表时间: 2019
• 期刊: Journal Of Human Behavior In The Social Environment
• 影响因子: 1.6
• 作者: Stover,AmandaN;Winstanley,ErinL;Zhang,Yifan;Feinberg,Judith
• 通讯作者: Feinberg,Judith

Contrast Feature Dependency Pattern Mining for Controlled Experiments with Application to Driving Behavior

用于受控实验的对比特征依赖模式挖掘及其在驾驶行为中的应用

• DOI: 10.1109/icdm.2019.00146
• 发表时间: 2019
• 期刊: 19th IEEE International Conference on Data Mining (ICDM
• 作者: Li, Qingzhe;Zhao, Liang;Lee, Yi-Ching;Ye, Yanfang;Lin, Jessica;Wu, Lingfei
• 通讯作者: Wu, Lingfei

Accelerating Convolutional Neural Network via Structured Gaussian Scale Mixture Models: A Joint Grouping and Pruning Approach

• DOI: 10.1109/jstsp.2020.2975610
• 发表时间: 2020-02
• 期刊: IEEE Journal of Selected Topics in Signal Processing
• 影响因子: 7.5
• 作者: Tao Huang;W. Dong;Jinshan Liu;Fangfang Wu;Guangming Shi;Xin Li