Artificial intelligence approaches using natural language processing to advance EHR-based clinical research

doi:10.1016/j.jaci.2019.12.897

Journal of Allergy and Clinical Immunology

Volume 145, Issue 2, February 2020, Pages 463-469

https://doi.org/10.1016/j.jaci.2019.12.897 Get rights and content

The wide adoption of electronic health record systems in health care generates big real-world data that open new venues to conduct clinical research. As a large amount of valuable clinical information is locked in clinical narratives, natural language processing techniques as an artificial intelligence approach have been leveraged to extract information from clinical narratives in electronic health records. This capability of natural language processing potentially enables automated chart review for identifying patients with distinctive clinical characteristics in clinical care and reduces methodological heterogeneity in defining phenotype, obscuring biological heterogeneity in research concerning allergy, asthma, and immunology. This brief review discusses the current literature on the secondary use of electronic health record data for clinical research concerning allergy, asthma, and immunology and highlights the potential, challenges, and implications of natural language processing techniques.

Section snippets

Systematic review for NLP-based research in allergy, asthma, and immunology

To assess the current state of EHR-based research using NLP in the field of asthma, allergy, and immunology, a comprehensive literature search of several databases from January 1, 2000, to August 13, 2019, English language, was conducted (see Table E1 in this article’s Online Repository at www.jacionline.org for the detailed method for our systematic literature search and the summary of results of each study). Only 21 articles were included in this systematic review by excluding abstracts,

Introduction for NLP

We discuss a brief overview of NLP to introduce the conceptual understanding of NLP for clinicians but do not intend to cover the topic of NLP for the purpose of performing NLP-based research. We refer readers to a few review articles that discuss NLP in depth.38, 39, 40 NLP is a field in computer science, AI, and computational linguistics that enables interactions between computers and human languages and bridges the gap between clinical human language and computational systems. NLP, ML, and

Implications of NLP for EHR based on clinical research and care

Fig 2 illustrates modular components of EHRs (top row) that encompass specific clinical information in both structured (eg, laboratory data) and nonstructured (eg, clinical notes) formats (bottom row), and the multimodal capability of NLP can extract and process this information, and classify patients, enabling important clinical tasks (right column) such as quality report and clinical decision support. Thus, NLP might be an important method to make EHRs a helpful data source for addressing the

References (53)

Y. Wang et al.
Clinical information extraction applications: a literature review
J Biomed Inform
(2018)
S.T. Wu et al.
Automated chart review for asthma cohort identification using natural language processing: an exploratory study
Ann Allergy Asthma Immunol
(2013)
S. Wu et al.
Modeling asynchronous event sequences with RNNs
J Biomed Inform
(2018)
C.I. Wi et al.
Natural language processing for asthma ascertainment in different practice settings
J Allergy Clin Immunol Pract
(2018)
S. Sohn et al.
Ascertainment of asthma prognosis using natural language processing from electronic medical records
J Allergy Clin Immunol
(2018)
K.H. Lai et al.
Automated misspelling detection and correction in clinical free-text records
J Biomed Inform
(2015)
I. Segura-Bedmar et al.
Predicting of anaphylaxis in big data EMR by exploring machine learning approaches
J Biomed Inform
(2018)
K.G. Blumenthal et al.
Adverse and hypersensitivity reactions to prescription nonsteroidal anti-inflammatory agents in a large health care system
J Allergy Clin Immunol Pract
(2017)
C. Friedman et al.
Natural language processing: state of the art and prospects for significant progress, a workshop sponsored by the National Library of Medicine
J Biomed Inform
(2013)
W.H. Crown
Potential application of machine learning in health outcomes research and some statistical cautions
Value Health
(2015)

M. Topalovic et al.

Pulmonary Function Study Investigators; Pulmonary Function Study Investigators:. Artificial intelligence outperforms pulmonologists in the interpretation of pulmonary function tests

Eur Respir J

(2019)

Z.I. Attia et al.

Screening for cardiac contractile dysfunction using an artificial intelligence–enabled electrocardiogram

Nat Med

(2019)

E. Mercan et al.

Assessment of machine learning of breast pathology structures for automated differentiation of breast cancer and high-risk proliferative lesions

JAMA Netw Open

(2019)

National Institutes of Health. Clinical and Translational Science Awards (CTSA): Foundation for Accelerated Discovery...

A.A. Lemke et al.

Community engagement in biobanking: experiences from the eMERGE Network

Genomics Soc Policy

(2010)

R.L. Fleurence et al.

Launching PCORnet, a national patient-centered clinical research network

J Am Med Inform Assoc

(2014)

R.L. Gardner et al.

Physician stress and burnout: the impact of health information technology

J Am Med Inform Assoc

(2019)

EHRs should be a tool not a task. 2019

F. Martin-Sanchez et al.

Big data in medicine is driving big changes

Yearb Med Inform

(2014)

National NLP Clinical Challenges (n2c2). 2019

Y. Xiong et al.

Cohort selection for clinical trials using hierarchical neural network

J Am Med Inform Assoc

(2019)

G. Fang et al.

Applying machine learning to predict real-world individual treatment effects: insights from a virtual patient cohort

J Am Med Inform Assoc

(2019)

L. Chen et al.

Clinical trial cohort selection based on multi-level rule-based natural language processing system

J Am Med Inform Assoc

(2019)

J.P. Jarow et al.

Multidimensional evidence generation and FDA regulatory decision making: defining and using “real-world” data

JAMA

(2017)

C. Shivade et al.

A review of approaches to identifying patient phenotype cohorts using electronic health records

J Am Med Inform Assoc

(2014)

S.T. Wu et al.

Patient-level temporal aggregation for text-based asthma status ascertainment

J Am Med Inform Assoc

(2014)

Cited by (127)

Learning medical concept representation based on semantic information in medical textural data
2024, Expert Systems with Applications
Electric Health Records (EHR) have been widely adopted by many hospitals to improve clinical decision making and re-admission prediction. Each patient admission usually contains both multiple medical codes as well as clinical notes. Accurate learning of the representations (also called embeddings) of medical concepts from EHR is a key strategy to improve prediction performance in healthcare. Existing works employ medical ontologies to improve the quality of representations but focus solely on the relationships amongst the medical codes, ignoring textual data such medical code descriptions, clinical notes, and patient demographics. In this paper, we propose a new model called Semantic-based Attention model using Textual data for Medical Concept Embedding (SATexMCE). SATexMCE consists of three parts: medical codes embedding, admission embedding, and prediction model. First, we generate representations of medical codes by using both the textual description of medical codes and also the relationships among medical codes. Then, we generate the admission representation based on the representation of medical codes in the admission, the clinical notes and the demographic information associated with the admission via several attention mechanisms. The admission embeddings are used to construct a Recurrent neural network model which is used to predict patients’ readmission and a disease in the next admission based on patient admission data. Experimental results show that the proposed SATexMCE model improves not only the performance of readmission prediction but also the quality of medical concept representations. Attention mechanism helps us measure the importance of different medical codes and understand the meaning of the words in clinical notes for predictions.
Artificial intelligence in healthcare institutions: A systematic literature review on influencing factors
2024, Technology in Society
The purpose of this review is integrating and contextualizing relevant literature on the factors influencing the adoption of AI in the healthcare industry into a comprehensive framework. Health systems are considered fundamental to creating societal value. However, global health systems are challenged by the increasing number of patients due to population aging and the growing prevalence of chronic diseases and cancer. Meanwhile, the United Nations calls for equal access to healthcare, tackling costs, and addressing resource constraints to foster the sustainable development of societies. In this context, artificial intelligence (AI) is gaining attention as it constitutes a promising technology to address these burgeoning challenges. Despite opportunities, the literature specifically on the adoption of AI in the healthcare industry is fragmented across various research fields, lacking a comprehensive overview. It lacks theoretically grounded research integrating, for example, the factors that influence the adoption of AI in healthcare institutions.
Derived from a multi-disciplinary systematic literature review, building on 130 studies, we propose the Adoption of AI in the Healthcare Industry Model. This model encompasses five dimensions influencing the adoption of AI in the healthcare industry and contextualizes them. We propose that macro-economic, regulatory, and technological readiness serve as external antecedents whereas organizational and individual readiness constitute internal antecedents influencing adoption of AI in healthcare institutions.
Our review has implications for research on technology acceptance related to AI in healthcare. Further, we provide hands-on guidance for AI providers, health institutions, and official bodies such as governments to foster the adoption of AI to leverage value.
Rural Health Disparities in Allergy, Asthma, and Immunologic Diseases: The Current State and Future Direction for Clinical Care and Research
2024, Journal of Allergy and Clinical Immunology: In Practice
Rural health disparities are well documented and continue to jeopardize the long-term health and wellness for the millions of individuals who live in rural America. The disparities observed between urban and rural residents encompass numerous morbidity and mortality measures for several chronic diseases and have been referred to as the “rural mortality penalty.” Although the unmet health needs of rural communities are widely acknowledged, little is known about rural health disparities in allergies, asthma, and immunologic diseases. Furthermore, the intersection between rural health disparities and social determinants of health has not been fully explored. To achieve a more complete understanding of the factors that perpetuate rural health disparities, greater research efforts followed by improved practice and policy are needed that account for the complex social context within rural communities rather than a general comparison between urban and rural environments or focusing on biomedical factors. Moreover, research efforts must prioritize community inclusion throughout rural areas through meaningful engagement of stakeholders in both clinical care and research. In this review, we examine the scope of health disparities in the rural United States and the impact of social determinants of health. We then detail the current state of rural health disparities in the field of allergy, asthma, and immunology. To close, we offer future considerations to address knowledge gaps and unmet needs for both clinical care and research in addressing rural health disparities.
TGRA-P: Task-driven model predicts 90-day mortality from ICU clinical notes on mechanical ventilation
2023, Computer Methods and Programs in Biomedicine
With the outbreak and spread of COVID-19 worldwide, limited ventilators fail to meet the surging demand for mechanical ventilation in the ICU. Clinical models based on structured data that have been proposed to rationalize ventilator allocation often suffer from poor ductility due to fixed fields and laborious normalization processes. The advent of pre-trained models and downstream fine-tuning methods allows for learning large amounts of unstructured clinical text for different tasks. But the hardware requirements of large-scale pre-trained models and purposeless networks downstream have led to a lack of promotion in the clinical domain.
In this study, an innovative architecture of a task-driven predictive model is proposed and a Task-driven Gated Recurrent Attention Pool model (TGRA-P) is developed based on the architecture. TGRA-P predicts early mortality risk from patients' clinical notes on mechanical ventilation in the ICU, which is used to assist clinicians in diagnosis and decision-making.
Specifically, a Task-Specific Embedding Module is proposed to fine-tune the embedding with task labels and save it as static files for downstream calls. It serves the task better and prevents GPU overload. The Gated Recurrent Attention Unit (GRA) is proposed to further enhance the dependency of the information preceding and following the text sequence with fewer parameters. In addition, we propose a Residual Max Pool (RMP) to avoid ignoring words in common text classification tasks by incorporating all word-level features of the notes for prediction. Finally, we use a fully connected decoding network as a classifier to predict the mortality risk.
The proposed model shows very promising results with an AUROC of 0.8245±0.0096, an AUPRC of 0.7532±0.0115, an accuracy of 0.7422±0.0028 and F1-score of 0.6612±0.0059 for 90-day mortality prediction using clinical notes of ICU mechanically ventilated patients on the MIMIC-III dataset, all of which are better than previous studies. Moreover, the superiority of the proposed model in comparison with other baseline models is also statistically validated through the calculated Cohen's d effect sizes.
The experimental results show that TGRA-P based on the innovative task-driven prognostic architecture obtains state-of-the-art performance. In future work, we will build upon the provided code and investigate its applicability to different datasets. The model balances performance and efficiency, not only reducing the cost of early mortality risk prediction but also assisting physicians in making timely clinical interventions and decisions. By incorporating textual records that are challenging for clinicians to utilize, the model serves as a valuable complement to physicians' judgment, enhancing their decision-making process.
Predicting post–liver transplant outcomes in patients with acute-on-chronic liver failure using Expert-Augmented Machine Learning
2023, American Journal of Transplantation
Liver transplantation (LT) is a treatment for acute-on-chronic liver failure (ACLF), but high post-LT mortality has been reported. Existing post-LT models in ACLF have been limited. We developed an Expert-Augmented Machine Learning (EAML) model to predict post-LT outcomes. We identified ACLF patients who underwent LT in the University of California Health Data Warehouse. We applied the RuleFit machine learning (ML) algorithm to extract rules from decision trees and create intermediate models. We asked human experts to rate the rules generated by RuleFit and incorporated these ratings to generate final EAML models. We identified 1384 ACLF patients. For death at 1 year, areas under the receiver-operating characteristic curve were 0.707 (confidence interval [CI] 0.625-0.793) for EAML and 0.719 (CI 0.640-0.800) for RuleFit. For death at 90 days, areas under the receiver-operating characteristic curve were 0.678 (CI 0.581-0.776) for EAML and 0.707 (CI 0.615-0.800) for RuleFit. In pairwise comparisons, both EAML and RuleFit models outperformed cross-sectional models. Significant discrepancies between experts and ML occurred in rankings of biomarkers used in clinical practice. EAML may serve as a method for ML-guided hypothesis generation in further ACLF research.
Application of Natural Language Processing in Total Joint Arthroplasty: Opportunities and Challenges
2023, Journal of Arthroplasty
Total joint arthroplasty is becoming one of the most common surgeries within the United States, creating an abundance of analyzable data to improve patient experience and outcomes. Unfortunately, a large majority of this data is concealed in electronic health records only accessible by manual extraction, which takes extensive time and resources. Natural language processing (NLP), a field within artificial intelligence, may offer a viable alternative to manual extraction. Using NLP, a researcher can analyze written and spoken data and extract data in an organized manner suitable for future research and clinical use. This article will first discuss common subtasks involved in an NLP pipeline, including data preparation, modeling, analysis, and external validation, followed by examples of NLP projects. Challenges and limitations of NLP will be discussed, closing with future directions of NLP projects, including large language models.

View all citing articles on Scopus

: The work was supported by grants from the National Institute of Health (grant no. R01 HL126667) and the R21 grant (grant no. R21AI116839-01).

: Disclosure of potential conflict of interest: The authors declare that they have no relevant conflicts of interest.

View full text

Reviews and feature articleArtificial intelligence approaches using natural language processing to advance EHR-based clinical research

Section snippets

Systematic review for NLP-based research in allergy, asthma, and immunology

Introduction for NLP

Implications of NLP for EHR based on clinical research and care

J Biomed Inform

Ann Allergy Asthma Immunol

J Biomed Inform

J Allergy Clin Immunol Pract

J Allergy Clin Immunol

J Biomed Inform

J Biomed Inform

J Allergy Clin Immunol Pract

J Biomed Inform

Value Health

Pulmonary Function Study Investigators; Pulmonary Function Study Investigators:. Artificial intelligence outperforms pulmonologists in the interpretation of pulmonary function tests

Eur Respir J

Screening for cardiac contractile dysfunction using an artificial intelligence–enabled electrocardiogram

Nat Med

Assessment of machine learning of breast pathology structures for automated differentiation of breast cancer and high-risk proliferative lesions

JAMA Netw Open

Community engagement in biobanking: experiences from the eMERGE Network

Genomics Soc Policy

Launching PCORnet, a national patient-centered clinical research network

J Am Med Inform Assoc

Physician stress and burnout: the impact of health information technology

J Am Med Inform Assoc

EHRs should be a tool not a task. 2019

Big data in medicine is driving big changes

Yearb Med Inform

National NLP Clinical Challenges (n2c2). 2019

Cohort selection for clinical trials using hierarchical neural network

J Am Med Inform Assoc

Applying machine learning to predict real-world individual treatment effects: insights from a virtual patient cohort

J Am Med Inform Assoc

Clinical trial cohort selection based on multi-level rule-based natural language processing system

J Am Med Inform Assoc

Multidimensional evidence generation and FDA regulatory decision making: defining and using “real-world” data

JAMA

A review of approaches to identifying patient phenotype cohorts using electronic health records

J Am Med Inform Assoc

Patient-level temporal aggregation for text-based asthma status ascertainment

J Am Med Inform Assoc

Reviews and feature article
Artificial intelligence approaches using natural language processing to advance EHR-based clinical research