Identification of asthma control factor in clinical notes using a hybrid deep learning model.
Identification of asthma control factor in clinical notes using a hybrid deep learning model.
复制标题
DOI：
10.1186/s12911-021-01633-4
复制
发表时间：
2021-11-09
影响因子：
3.5
通讯作者：
Sohn S
中科院分区：
文献类型：
--
作者：
Agnikula Kshatriya BS;Sagheb E;Wi CI;Yoon J;Seol HY;Juhn Y;Sohn S
哮喘护理中符合指南的文件存在显着差异。然而，仅使用结构化数据评估临床医生的记录是不可行的，而是需要对电子健康记录 (EHR) 进行劳动密集型图表审查。哮喘控制因素中的某些指导元素（例如审查吸入器技术）需要理解上下文才能从 EHR 自由文本中正确捕获。研究数据由两组组成：(1) 手动图表审查数据——300 名哮喘诊断患者的 1039 份临床记录，(2) 弱标记数据（远程监督）——800 名哮喘诊断患者的 27,363 份临床记录。开发了上下文感知语言模型 Transformers 双向编码器表示 (BERT)，用于识别 EHR 自由文本中的吸入器技术。原始 BERT 和临床 BioBERT (cBERT) 都以成本敏感性的方式应用来处理不平衡数据。还结合了按规则使用弱标签的远程监督，以增强训练集并减轻深度学习算法开发中昂贵的手动标记过程。还探索了使用事后规则的混合方法来修复 BERT 模型错误。在精度、召回率、F 分数和准确性方面比较了有/无远程监督、混合模型和基于规则的模型的 BERT 性能。原始数据上的 BERT 模型的 F1 分数与基于规则的模型类似（规则、BERT 和 cBERT 分别为 0.837、0.845 和 0.838）。与没有远程监督和基于规则的模型相比，具有远程监督的 BERT 模型产生了更高的性能（BERT 和 cBERT 分别为 0.853 和 0.880）。混合模型在 BERT 和 cBERT 的远程监督中表现最好，F1 分数为 0.877 和 0.904。所提出的具有远程监督的 BERT 模型证明了其在 EHR 自由文本中识别吸入器技术的能力，并且优于基于规则的模型和在原始数据上训练的 BERT 模型。通过远程监督方法，我们可以减少昂贵的手动图表审查，以生成大多数基于深度学习的模型所需的大量训练数据。混合模型能够修复 BERT 模型错误并进一步提高性能。
There are significant variabilities in guideline-concordant documentation in asthma care. However, assessing clinician’s documentation is not feasible using only structured data but requires labor-intensive chart review of electronic health records (EHRs). A certain guideline element in asthma control factors, such as review inhaler techniques, requires context understanding to correctly capture from EHR free text. The study data consist of two sets: (1) manual chart reviewed data—1039 clinical notes of 300 patients with asthma diagnosis, and (2) weakly labeled data (distant supervision)—27,363 clinical notes from 800 patients with asthma diagnosis. A context-aware language model, Bidirectional Encoder Representations from Transformers (BERT) was developed to identify inhaler techniques in EHR free text. Both original BERT and clinical BioBERT (cBERT) were applied with a cost-sensitivity to deal with imbalanced data. The distant supervision using weak labels by rules was also incorporated to augment the training set and alleviate a costly manual labeling process in the development of a deep learning algorithm. A hybrid approach using post-hoc rules was also explored to fix BERT model errors. The performance of BERT with/without distant supervision, hybrid, and rule-based models were compared in precision, recall, F-score, and accuracy. The BERT models on the original data performed similar to a rule-based model in F1-score (0.837, 0.845, and 0.838 for rules, BERT, and cBERT, respectively). The BERT models with distant supervision produced higher performance (0.853 and 0.880 for BERT and cBERT, respectively) than without distant supervision and a rule-based model. The hybrid models performed best in F1-score of 0.877 and 0.904 over the distant supervision on BERT and cBERT. The proposed BERT models with distant supervision demonstrated its capability to identify inhaler techniques in EHR free text, and outperformed both the rule-based model and BERT models trained on the original data. With a distant supervision approach, we may alleviate costly manual chart review to generate the large training data required in most deep learning-based models. A hybrid model was able to fix BERT model errors and further improve the performance.
登录
查看更多内容
影响因子：
3.1
作者：
Kaur H;Sohn S;Wi CI;Ryu E;Park MA;Bachman K;Kita H;Croghan I;Castro-Rodriguez JA;Voge GA;Liu H;Juhn YJ
通讯作者：
Juhn YJ
DOI：
10.1016/j.jaci.2019.12.897
发表时间：
2020-02
期刊：
The Journal of allergy and clinical immunology
影响因子：
--
作者：
Juhn Y;Liu H
通讯作者：
Liu H
影响因子：
9.6
作者：
Senthilselvan, A;Lawson, JA;Dosman, JA
通讯作者：
Dosman, JA
影响因子：
5.1
作者：
Cloutier, MM;Hall, CB;Bailit, H
通讯作者：
Bailit, H
影响因子：
14.2
作者：
Sohn, Sunghwan;Wi, Chung-Il;Juhn, Young J.
通讯作者：
Juhn, Young J.