The ngram chief complaint classifier: A novel method of automatically creating chief complaint classifiers based on international classification of diseases groupings

Brown, Philip E.; Halász, Sylvia; Goodall, Colin; Cochrane, Dennis G.; Milano, Peter M.

doi:10.1016/j.jbi.2009.08.015

Cited by 11 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since our automatic syndrome classifier uses the data available during the very initial stages of the patient's ED visit, it is safe to assume that we can obtain results in a timely manner. Almost all of the existing syndrome classification systems are highly specific; for example, the ngram classifier [3], CoCo [8] and EARS [14] have a 95% or higher specificity at the cost of sensitivity (50%-75%). We investigated the reason for the acceptance of lower sensitivity values by varying the threshold value of cosine similarity for the entire test set using both triage note and chief complaint.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Automated syndrome classification using early phase emergency department data

Mahalingam

Mostafa

Travers

et al. 2012

Proceedings of the 2nd ACM SIGHIT International Health Informatics Symposium

View full text Add to dashboard Cite

The primary motivation behind automated syndrome classification is to shorten the time it may take to detect an outbreak or a community-wide public health issue. State-of-the-art syndrome classification techniques have primarily used Emergency Department (ED) chief complaints (CCs) or other short free text descriptions of patient symptoms for near real-time surveillance purposes. We propose a system that can automatically classify an ED record with initial temperature, chief complaint and triage nurse's notes into one or more syndromes using the vector space model. Terms from syndrome definitions and a small tf-idf weighted syndrome-positive training set are used to create a reference dictionary for generating the syndrome and triage note vectors. Subsequently, cosine similarity between the vectors is used to establish the particular syndrome category. We tested the system on a manually classified dataset of 485 ED records for the gastro-intestinal (GI) syndrome and measured performance in terms of sensitivity (~93%) and specificity (~82%). Initial results show that this system is capable of producing a better balance between these measures compared to existing automatic syndrome categorization systems which use only chief complaints and/or diagnosis codes.

show abstract

Section: Discussionmentioning

confidence: 99%

“…For example, the ngram chief complaint classifier [3] first builds a chief complaint classifier based on the classification of a training set using a simple ICD-9-CM classifier (ESSENCE). Both these classifiers were then used on the test set to determine if the visits belonged to Gastrointestinal (GI) syndrome or not.…”

Section: Introductionmentioning

confidence: 99%

Automated syndrome classification using early phase emergency department data

Mahalingam

Mostafa

Travers

et al. 2012

Proceedings of the 2nd ACM SIGHIT International Health Informatics Symposium

View full text Add to dashboard Cite

show abstract

“…The NC DETECT syndromes include standalone exclusion criteria for each of the syndromes [7]. For example, the gastrointestinal syndrome excludes records with “Crohn” and “irritable bowel” since these chronic conditions are associated with symptoms that are identical to many of those in the GI syndrome, which could lead to false positive classification.…”

Section: Methodsmentioning

confidence: 99%

“…For syndromic surveillance, "timeliness" is defined in terms of hours, and action by public health officials may be warranted before there is a definitive test result or diagnosis. Pre-diagnostic data used for surveillance include initial vital signs (e.g., measured temperature, heart rate) and chief complaint (CC) [7-9]. Researchers have also explored the addition of different portions of the electronic medical record to capture data in addition to chief complaint, such as discharge prescriptions, diagnostic test orders, structured clinical notes, and triage nurses' notes in narrative form [10-13].…”

Section: Introductionmentioning

confidence: 99%

“…Meystre et al [16] discuss some established techniques for information extraction from clinical unstructured text, such as incorporating supplemental information sources (e.g., the UMLS), mapping terms to a standard representation (e.g., ICD-9-CM codes), and cleaning and normalizing text using pre-processing software (e.g., Chief Complaint Processor (CCP), Emergency Medical Text Processor (EMT-P), NegEx, ConText) [3,7,17-27]. Although structured data would be easier to work with, fields such as the triage note are usually recorded in unstructured text.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Emergency Medical Text Classifier: New system improves processing and classification of triage notes

Haas

Travers

Waller

et al. 2014

OJPHI

View full text Add to dashboard Cite

ObjectiveAutomated syndrome classification aims to aid near real-time syndromic surveillance to serve as an early warning system for disease outbreaks, using Emergency Department (ED) data. We present a system that improves the automatic classification of an ED record with triage note into one or more syndrome categories using the vector space model coupled with a ‘learning’ module that employs a pseudo-relevance feedback mechanism. Materials and Methods: Terms from standard syndrome definitions are used to construct an initial reference dictionary for generating the syndrome and triage note vectors. Based on cosine similarity between the vectors, each record is classified into a syndrome category. We then take terms from the top-ranked records that belong to the syndrome of interest as feedback. These terms are added to the reference dictionary and the process is repeated to determine the final classification. The system was tested on two different datasets for each of three syndromes: Gastro-Intestinal (GI), Respiratory (Resp) and Fever-Rash (FR). Performance was measured in terms of sensitivity (Se) and specificity (Sp). Results: The use of relevance feedback produced high values of sensitivity and specificity for all three syndromes in both test sets: GI: 90% and 71%, Resp: 97% and 73%, FR: 100% and 87%, respectively, in test set 1, and GI: 88% and 69%, Resp: 87% and 61%, FR: 97% and 71%, respectively, in test set 2. Conclusions: The new system for pre-processing and syndromic classification of ED records with triage notes achieved improvements in Se and Sp. Our results also demonstrate that the system can be tuned to achieve different levels of performance based on user requirements.

show abstract

A practical tool for public health surveillance: Semi-automated coding of short injury narratives from large administrative databases using Naïve Bayes algorithms

Marucci‐Wellman

Lehto

Corns

2015

Accident Analysis & Prevention

View full text Add to dashboard Cite

Public health surveillance programs in the U.S. are undergoing landmark changes with the availability of electronic health records and advancements in information technology. Injury narratives gathered from hospital records, workers compensation claims or national surveys can be very useful for identifying antecedents to injury or emerging risks. However, classifying narratives manually can become prohibitive for large datasets. The purpose of this study was to develop a human-machine system that could be relatively easily tailored to routinely and accurately classify injury narratives from large administrative databases such as workers compensation. We used a semi-automated approach based on two Naïve Bayesian algorithms to classify 15,000 workers compensation narratives into two-digit Bureau of Labor Statistics (BLS) event (leading to injury) codes. Narratives were filtered out for manual review if the algorithms disagreed or made weak predictions. This approach resulted in an overall accuracy of 87%, with consistently high positive predictive values across all two-digit BLS event categories including the very small categories (e.g., exposure to noise, needle sticks). The Naïve Bayes algorithms were able to identify and accurately machine code most narratives leaving only 32% (4853) for manual review. This strategy substantially reduces the need for resources compared with manual review alone.

show abstract

The ngram chief complaint classifier: A novel method of automatically creating chief complaint classifiers based on international classification of diseases groupings

Abstract: The ngram CC classifier performed similarly to manually developed CC classifiers and has advantages of rapid automated creation and updating, and may be used independent of language or dialect.

Cited by 11 publications

References 17 publications

Automated syndrome classification using early phase emergency department data

Automated syndrome classification using early phase emergency department data

Emergency Medical Text Classifier: New system improves processing and classification of triage notes

A practical tool for public health surveillance: Semi-automated coding of short injury narratives from large administrative databases using Naïve Bayes algorithms

Contact Info

Product

Resources

About