Leveraging graph topology and semantic context for pharmacovigilance through twitter-streams.
Leveraging graph topology and semantic context for pharmacovigilance through twitter-streams.
复制标题
DOI：
10.1186/s12859-016-1220-5
复制
发表时间：
2016-10-06
影响因子：
3
通讯作者：
Singh R
中科院分区：
文献类型：
--
作者：
Eshleman R;Singh R
药物不良事件（ADE）是治疗后死亡的主要原因之一，其识别构成了现代精准医学的重要挑战。不幸的是，ADE 的发生和影响常常被低估，这使得及时干预变得复杂。 Twitter 每天发布超过 5 亿条帖子，是一个常用的社交媒体平台。 Twitter 上日常个人信息交换的普遍性使其成为用于 ADE 识别和干预的数据挖掘的有希望的目标。该问题的核心是三个技术挑战：(1) 识别（嘈杂的）推文中的显着医学关键词，(2) 绘制药效关系，以及 (3) 将此类关系分类为不良或非不良。我们使用称为药效图 (DEG) 的二部图论表示，通过将药物和副作用表示为顶点来对药物和副作用关系进行建模。我们在两个数据源上构建单独的 DEG。第一个 DEG 是根据 SIDER 数据库中记录的 FDA 说明书中发现的药效关系构建的。第二个DEG是通过挖掘Twitter用户的历史构建的。我们使用基于字典的信息提取来识别推文中与医学相关的概念。药物以及同时出现的症状与按时间距离和频率加权的边缘相关。最后，来自 SIDER DEG 的信息与 Twitter DEG 集成，并使用监督机器学习将边缘分类为不利或非不利。我们检查分类任务的图论和语义特征。所提出的方法可以高精度地识别药物不良反应，精度超过 85%，F1 超过 81%。与仅采用未丰富的图论分析的最先进的领先方法相比，我们的方法在上述措施方面取得了 5% 至 8% 的改进。此外，我们采用我们的方法发现了几种 ADE，尽管这些 ADE 存在于医学文献和 Twitter 流中，但并未出现在 SIDER 数据库中。我们提出了 DEG 集成模型，作为分析药效关系的强大形式主义，该模型足够通用，可以容纳不同的数据源，但足够严格，可以为 ADE 识别提供强大的机制。本文的在线版本 (doi:10.1186/s12859-016-1220-5) 包含补充材料，可供授权用户使用。
Adverse drug events (ADEs) constitute one of the leading causes of post-therapeutic death and their identification constitutes an important challenge of modern precision medicine. Unfortunately, the onset and effects of ADEs are often underreported complicating timely intervention. At over 500 million posts per day, Twitter is a commonly used social media platform. The ubiquity of day-to-day personal information exchange on Twitter makes it a promising target for data mining for ADE identification and intervention. Three technical challenges are central to this problem: (1) identification of salient medical keywords in (noisy) tweets, (2) mapping drug-effect relationships, and (3) classification of such relationships as adverse or non-adverse. We use a bipartite graph-theoretic representation called a drug-effect graph (DEG) for modeling drug and side effect relationships by representing the drugs and side effects as vertices. We construct individual DEGs on two data sources. The first DEG is constructed from the drug-effect relationships found in FDA package inserts as recorded in the SIDER database. The second DEG is constructed by mining the history of Twitter users. We use dictionary-based information extraction to identify medically-relevant concepts in tweets. Drugs, along with co-occurring symptoms are connected with edges weighted by temporal distance and frequency. Finally, information from the SIDER DEG is integrate with the Twitter DEG and edges are classified as either adverse or non-adverse using supervised machine learning. We examine both graph-theoretic and semantic features for the classification task. The proposed approach can identify adverse drug effects with high accuracy with precision exceeding 85 % and F1 exceeding 81 %. When compared with leading methods at the state-of-the-art, which employ un-enriched graph-theoretic analysis alone, our method leads to improvements ranging between 5 and 8 % in terms of the aforementioned measures. Additionally, we employ our method to discover several ADEs which, though present in medical literature and Twitter-streams, are not represented in the SIDER databases. We present a DEG integration model as a powerful formalism for the analysis of drug-effect relationships that is general enough to accommodate diverse data sources, yet rigorous enough to provide a strong mechanism for ADE identification. The online version of this article (doi:10.1186/s12859-016-1220-5) contains supplementary material, which is available to authorized users.
登录
查看更多内容
影响因子：
2.9
作者：
Bate, A;Lindquist, M;De Freitas, RM
通讯作者：
De Freitas, RM
影响因子：
8.3
作者：
FLETCHER, AE;BULPITT, CJ;NEISS, AM
通讯作者：
NEISS, AM
影响因子：
4.2
作者：
Heeley, E;Wilton, LV;Shakir, SAW
通讯作者：
Shakir, SAW
影响因子：
4.2
作者：
Freifeld, Clark C.;Brownstein, John S.;Menone, Christopher M.;Bao, Wenjie;Filice, Ross;Kass-Hout, Taha;Dasgupta, Nabarun
通讯作者：
Dasgupta, Nabarun
影响因子：
3
作者：
Harpaz R;Chase HS;Friedman C
通讯作者：
Friedman C