Natural language processing: an introduction

Nadkarni, Prakash M.; Ohno-Machado, Lucila; Chapman, Wendy W.

doi:10.1136/amiajnl-2011-000464

Cited by 1,230 publications

(634 citation statements)

References 42 publications

Supporting

Mentioning

623

Contrasting

Unclassified

Order By: Relevance

“…So, this approach can be generalized. [41][42][43] More broadly, this work highlights one way in which health information technology may facilitate process and quality improvement work. 44,45 This model's low sensitivity, high specificity, and high NPV make it well suited to help medical centers more effectively deploy resources to improve colonoscopy adherence.…”

Section: Discussionmentioning

confidence: 99%

Predicting Non-Adherence with Outpatient Colonoscopy Using a Novel Electronic Tool that Measures Prior Non-Adherence

et al. 2015

View full text Add to dashboard Cite

BACKGROUND: Accurately predicting the risk of no-show for a scheduled colonoscopy can help target interventions to improve compliance with colonoscopy, and thereby reduce the disease burden of colorectal cancer and enhance the utilization of resources within endoscopy units. OBJECTIVES: We aimed to utilize information available in an electronic medical record (EMR) and endoscopy scheduling system to create a predictive model for no-show risk, and to simultaneously evaluate the role for natural language processing (NLP) in developing such a model. DESIGN: This was a retrospective observational study using discovery and validation phases to design a colonoscopy non-adherence prediction model. An NLPderived variable called the Non-Adherence Ratio ("NAR") was developed, validated, and included in the model. PARTICIPANTS: Patients scheduled for outpatient colonoscopy at an Academic Medical Center (AMC) that is part of a multi-hospital health system, 2009 to 2011, were included in the study. MAIN MEASURES: Odds ratios for non-adherence were calculated for all variables in the discovery cohort, and an Area Under the Receiver Operating Curve (AUC) was calculated for the final non-adherence prediction model. KEY RESULTS: The non-adherence model included six variables: 1) gender; 2) history of psychiatric illness, 3) NAR; 4) wait time in months; 5) number of prior missed endoscopies; and 6) education level. The model achieved discrimination in the validation cohort (AUC= =70.2 %). At a threshold non-adherence score of 0.46, the model's sensitivity and specificity were 33 % and 92 %, respectively. Removing the NAR from the model significantly reduced its predictive power (AUC = 64.3 %, difference = 5.9 %, p<0.001). CONCLUSIONS: A six-variable model using readily available clinical and demographic information demonstrated accuracy for predicting colonoscopy non-adherence. The NAR, a novel variable developed using NLP technology, significantly strengthened this model's predictive power.

show abstract

Section: Discussionmentioning

confidence: 99%

Predicting Non-Adherence with Outpatient Colonoscopy Using a Novel Electronic Tool that Measures Prior Non-Adherence

et al. 2015

View full text Add to dashboard Cite

show abstract

“…In the clinical domain, NLP has been utilized to extract relevant information such as laboratory results, medications, and diagnoses from de-identified medical patient record narratives in order to identify patient cohorts that fit eligibility criteria for clinical research studies [16]. When compared to human chart review of medical records, NLP yields faster results [17][18][19][20]. NLP techniques have also been used to identify possible lung cancer patients based on their radiology reports [21] and extract disease characteristics for prostate cancer patients [22].…”

mentioning

confidence: 99%

Improving Patient Cohort Identification Using Natural Language Processing

Sarmiento

Dernoncourt

2016

Secondary Analysis of Electronic Health Records

View full text Add to dashboard Cite

“…In the UK, there has been mandatory use of clinical coding standards for several decades which makes UK EHR data a particularly rich source for data mining [26]. An alternative approach is to use Natural Language Processing (NLP) [27][28][29] of the unstructured notes and this has been particularly important in countries such as the USA where coding standards have not been widely adopted.…”

Section: Context: Big Data and Electronic Health Recordsmentioning

confidence: 99%

NETIMIS: Dynamic Simulation of Health Economics Outcomes Using Big Data

2016

View full text Add to dashboard Cite

Many healthcare organizations are now making good use of e-health record (EHR) systems to record clinical information about their patients and the details of their healthcare. Electronic data in EHRs is generated by people engaged in complex processes within complex environments and their human input, albeit shaped by computer systems, is compromised by many human factors. This data is potentially valuable to health economists and outcomes researchers but is sufficiently large and complex enough to be considered part of the new frontier of "big data". This paper describes emerging methods that draw together data mining, process modeling, activity based costing and dynamic simulation models. Our research infrastructure includes safe links to Leeds hospital's EHR with 3 million secondary and tertiary care patients. We created a multi-disciplinary team of health economists, clinical specialists, data and computer scientists and developed a dynamic simulation tool called NETIMIS (www.netimis.com) suitable for visualization of both human-designed and data-mined processes which can then be used for "what-if" analysis by stakeholders interested in costing, designing and evaluating healthcare interventions. We present two examples of model development to illustrate how dynamic simulation can be informed by big data from an EHR. We found the tool provided a focal point for multi-disciplinary team work to help them iteratively and collaboratively "deep dive" into big data. Key Points for Decision Makers Big data can be defined as a data at a scale where research become uncomfortable -the point where existing methods are unable to unlock the value in the data.  Electronic health records (EHR) are a type of big data that will be valuable to future health economists and outcomes researchers but the data is messy, it is the human product of large numbers of individuals working in complex environments. Solutions have yet to emerge.  We found that the use of a dynamic simulation tool provided an effective focal point for multidisciplinary investigations into the messy world of EHR big data and the development of valuable insights.2

show abstract

Natural language processing: an introduction

Cited by 1,230 publications

References 42 publications

Predicting Non-Adherence with Outpatient Colonoscopy Using a Novel Electronic Tool that Measures Prior Non-Adherence

Predicting Non-Adherence with Outpatient Colonoscopy Using a Novel Electronic Tool that Measures Prior Non-Adherence

Improving Patient Cohort Identification Using Natural Language Processing

NETIMIS: Dynamic Simulation of Health Economics Outcomes Using Big Data

Contact Info

Product

Resources

About