A Deep Learning Method to Automatically Identify Reports of Scientifically Rigorous Clinical Research from the Biomedical Literature: Comparative Analytic Study.
A Deep Learning Method to Automatically Identify Reports of Scientifically Rigorous Clinical Research from the Biomedical Literature: Comparative Analytic Study.
复制标题
DOI：
10.2196/10281
复制
发表时间：
2018-06-25
影响因子：
7.4
通讯作者：
Haynes RB
中科院分区：
文献类型：
--
作者：
Del Fiol G;Michelson M;Iorio A;Cotoi C;Haynes RB
循证医学实践的一个主要障碍是有效地找到关于给定临床主题的科学合理的研究。研究一种深度学习方法，从生物医学文献中检索科学合理的治疗研究。我们使用403，216篇PubMed引文的噪声数据集训练卷积神经网络，其中标题和摘要作为特征。将深度学习模型与最先进的搜索过滤器进行比较，例如PubMed的临床查询广泛治疗过滤器，McMaster的文本搜索策略（无医学主题标题，MeSH，术语）和临床查询平衡治疗过滤器。先前注释的数据集（临床对冲）被用作金标准。深度学习模型获得的召回率显著低于临床广泛治疗过滤器（96.9% vs 98.4%; P<0.001）;与McMaster的文本搜索（96.9% vs 97.1%; P= 0.57）和临床广泛平衡过滤器（96.9% vs 97.0%; P= 0.63）相当。深度学习获得的精确度显著高于临床筛选器（34.6% vs 22.4%; P<.001）和McMaster的文本搜索（34.6% vs 11.8%; P<.001），但显著低于临床筛选器（34.6% vs 40.9%; P<.001）。与最先进的搜索过滤器相比，深度学习表现良好，尤其是在引文未被索引的情况下。与以前的机器学习方法不同，所提出的深度学习模型不需要特征工程，或时间敏感或专有特征，如MeSH术语和文献计量学。深度学习是一种很有前途的方法，可以识别科学严谨的临床研究报告。需要进一步的工作来优化深度学习模型，并评估对其他领域的可推广性，如诊断，病因和预后。
A major barrier to the practice of evidence-based medicine is efficiently finding scientifically sound studies on a given clinical topic. To investigate a deep learning approach to retrieve scientifically sound treatment studies from the biomedical literature. We trained a Convolutional Neural Network using a noisy dataset of 403,216 PubMed citations with title and abstract as features. The deep learning model was compared with state-of-the-art search filters, such as PubMed’s Clinical Query Broad treatment filter, McMaster’s textword search strategy (no Medical Subject Heading, MeSH, terms), and Clinical Query Balanced treatment filter. A previously annotated dataset (Clinical Hedges) was used as the gold standard. The deep learning model obtained significantly lower recall than the Clinical Queries Broad treatment filter (96.9% vs 98.4%; P<.001); and equivalent recall to McMaster’s textword search (96.9% vs 97.1%; P=.57) and Clinical Queries Balanced filter (96.9% vs 97.0%; P=.63). Deep learning obtained significantly higher precision than the Clinical Queries Broad filter (34.6% vs 22.4%; P<.001) and McMaster’s textword search (34.6% vs 11.8%; P<.001), but was significantly lower than the Clinical Queries Balanced filter (34.6% vs 40.9%; P<.001). Deep learning performed well compared to state-of-the-art search filters, especially when citations were not indexed. Unlike previous machine learning approaches, the proposed deep learning model does not require feature engineering, or time-sensitive or proprietary features, such as MeSH terms and bibliometrics. Deep learning is a promising approach to identifying reports of scientifically rigorous clinical research. Further work is needed to optimize the deep learning model and to assess generalizability to other areas, such as diagnosis, etiology, and prognosis.
登录
查看更多内容
DOI：
10.1197/jamia.m1641
发表时间：
2005-03-01
影响因子：
6.4
作者：
Aphinyanaphongs, Y;Tsamardinos, I;Aliferis, CF
通讯作者：
Aliferis, CF
DOI：
10.1197/jamia.m1909
发表时间：
2006-01-01
影响因子：
6.4
作者：
Bernstam, EV;Herskovic, JR;Hersh, WR
通讯作者：
Hersh, WR
影响因子：
15.8
作者：
Ioannidis, JPA
通讯作者：
Ioannidis, JPA
DOI：
10.1197/jamia.m2996
发表时间：
2009-01-01
影响因子：
6.4
作者：
Kilicoglu, Halil;Demner-Fushman, Dina;Haynes, R. Brian
通讯作者：
Haynes, R. Brian
影响因子：
3.5
作者：
Wilczynski, Nancy L;Morgan, Douglas;Haynes, R Brian
通讯作者：
Haynes, R Brian