Classifying Characteristics of Opioid Use Disorder From Hospital Discharge Summaries Using Natural Language Processing.
Classifying Characteristics of Opioid Use Disorder From Hospital Discharge Summaries Using Natural Language Processing.
复制标题
DOI：
10.3389/fpubh.2022.850619
复制
发表时间：
2022
影响因子：
5.2
通讯作者：
Mowery, Danielle L.
中科院分区：
文献类型：
--
作者：
Poulsen, Melissa N.;Freda, Philip J.;Troiani, Vanessa;Davoudi, Anahita;Mowery, Danielle L.
关键词：
阿片类药物使用障碍 (OUD) 在卫生系统中诊断不足，限制了使用电子健康记录 (EHR) 对 OUD 进行研究。医疗记录可以通过记录的 OUD 体征和症状以及社会风险和行为来丰富结构化 EHR 数据。为了大规模捕获这些信息，必须开发和评估自然语言处理 (NLP) 工具。我们开发并应用了注释模式来深入表征 OUD 以及相关的临床、行为和环境因素，并使用机器学习和基于深度学习的方法自动化注释模式。使用 MIMIC-III 重症监护数据库，我们查询了具有国际疾病分类 (ICD-9) OUD 诊断代码的患者的出院摘要。我们开发了一个注释模式来描述有问题的阿片类药物使用情况，识别具有潜在 OUD 的个体，并提供社会心理背景。两名注释者审查了 100 名患者的出院摘要。我们随机抽取患者及其相关注释句子，并将其分为训练组（66 名患者；2,127 个注释句子）和测试组（29 名患者；1,149 个注释句子）。我们使用训练集来生成特征，采用三种 NLP 算法/知识源。我们使用传统机器学习器（逻辑回归）和深度学习方法（基于 ELECTRA 替换的标记检测模型的 Autogluon）训练和测试分类预测模型。我们应用了五重交叉验证方法来减少性能估计中的偏差。生成的注释模式包含 32 个类。我们实现了适度的注释者间一致性，所有类别的 F1 分数从 48% 增加到 66%。五个类具有足够数量的自动化注释；其中，我们在药物筛选（训练：91-96；测试：91-94）和阿片类药物类型（训练：86-96；测试：86-99）的训练和测试集上观察到一致的高性能（F1 分数）。其他药物使用（训练：52-65；测试：40-48）、疼痛管理（训练：72-78；测试：61-78）和精神科（训练：73-80；测试：72）的训练和测试集表现均有所下降。 Autogluon 取得了最高的性能。这项试点研究表明，注释者可以手动识别有关有问题的阿片类药物使用的丰富信息。然而，更多的训练样本和特征将提高我们从临床文本（包括来自门诊环境的文本）中可靠地识别不太常见类别的能力。
Opioid use disorder (OUD) is underdiagnosed in health system settings, limiting research on OUD using electronic health records (EHRs). Medical encounter notes can enrich structured EHR data with documented signs and symptoms of OUD and social risks and behaviors. To capture this information at scale, natural language processing (NLP) tools must be developed and evaluated. We developed and applied an annotation schema to deeply characterize OUD and related clinical, behavioral, and environmental factors, and automated the annotation schema using machine learning and deep learning-based approaches. Using the MIMIC-III Critical Care Database, we queried hospital discharge summaries of patients with International Classification of Diseases (ICD-9) OUD diagnostic codes. We developed an annotation schema to characterize problematic opioid use, identify individuals with potential OUD, and provide psychosocial context. Two annotators reviewed discharge summaries from 100 patients. We randomly sampled patients with their associated annotated sentences and divided them into training (66 patients; 2,127 annotated sentences) and testing (29 patients; 1,149 annotated sentences) sets. We used the training set to generate features, employing three NLP algorithms/knowledge sources. We trained and tested prediction models for classification with a traditional machine learner (logistic regression) and deep learning approach (Autogluon based on ELECTRA's replaced token detection model). We applied a five-fold cross-validation approach to reduce bias in performance estimates. The resulting annotation schema contained 32 classes. We achieved moderate inter-annotator agreement, with F1-scores across all classes increasing from 48 to 66%. Five classes had a sufficient number of annotations for automation; of these, we observed consistently high performance (F1-scores) across training and testing sets for drug screening (training: 91–96; testing: 91–94) and opioid type (training: 86–96; testing: 86–99). Performance dropped from training and to testing sets for other drug use (training: 52–65; testing: 40–48), pain management (training: 72–78; testing: 61–78) and psychiatric (training: 73–80; testing: 72). Autogluon achieved the highest performance. This pilot study demonstrated that rich information regarding problematic opioid use can be manually identified by annotators. However, more training samples and features would improve our ability to reliably identify less common classes from clinical text, including text from outpatient settings.
登录
查看更多内容
影响因子：
2.6
作者：
Hazlehurst, Brian;Green, Carla A.;Coplan, Paul M.
通讯作者：
Coplan, Paul M.
影响因子：
9.8
作者：
Johnson AE;Pollard TJ;Shen L;Lehman LW;Feng M;Ghassemi M;Moody B;Szolovits P;Celi LA;Mark RG
通讯作者：
Mark RG
影响因子：
4.5
作者：
Chapman BE;Lee S;Kang HP;Chapman WW
通讯作者：
Chapman WW
影响因子：
3.1
作者：
Haller, Irina V.;Renier, Colleen M.;Elliott, Thomas E.
通讯作者：
Elliott, Thomas E.
影响因子：
3.5
作者：
Sharma, Brihat;Dligach, Dmitriy;Afshar, Majid
通讯作者：
Afshar, Majid