Comparison of UMLS terminologies to identify risk of heart disease using clinical notes.
Ontology highlight
ABSTRACT:
SUBMITTER: Shivade C
PROVIDER: S-EPMC4973866 | biostudies-other | 2015 Dec
REPOSITORIES: biostudies-other
Publications
Comparison of UMLS terminologies to identify risk of heart disease using clinical notes.
Shivade Chaitanya C Malewadkar Pranav P Fosler-Lussier Eric E Lai Albert M AM
Journal of biomedical informatics 20150912
The second track of the 2014 i2b2 challenge asked participants to automatically identify risk factors for heart disease among diabetic patients using natural language processing techniques for clinical notes. This paper describes a rule-based system developed using a combination of regular expressions, concepts from the Unified Medical Language System (UMLS), and freely-available resources from the community. With a performance (F1=90.7) that is significantly higher than the median (F1=87.20) an ...[more]
Similar Datasets
Project description:With increasing adoption of electronic health records (EHRs), there is an opportunity to use the free-text portion of EHRs for pharmacovigilance. We present novel methods that annotate the unstructured clinical notes and transform them into a deidentified patient-feature matrix encoded using medical terminologies. We demonstrate the use of the resulting high-throughput data for detecting drug-adverse event associations and adverse events associated with drug-drug interactions. We show that these methods flag adverse events early (in most cases before an official alert), allow filtering of spurious signals by adjusting for potential confounding, and compile prevalence information. We argue that analyzing large volumes of free-text clinical notes enables drug safety surveillance using a yet untapped data source. Such data mining can be used for hypothesis generation and for rapid analysis of suspected adverse event risk.
| S-EPMC3846296 | biostudies-literature
Project description:ObjectivesThere is much interest in utilizing clinical data for developing prediction models for Alzheimer's disease (AD) risk, progression, and outcomes. Existing studies have mostly utilized curated research registries, image analysis, and structured electronic health record (EHR) data. However, much critical information resides in relatively inaccessible unstructured clinical notes within the EHR.Materials and methodsWe developed a natural language processing (NLP)-based pipeline to extract AD-related clinical phenotypes, documenting strategies for success and assessing the utility of mining unstructured clinical notes. We evaluated the pipeline against gold-standard manual annotations performed by 2 clinical dementia experts for AD-related clinical phenotypes including medical comorbidities, biomarkers, neurobehavioral test scores, behavioral indicators of cognitive decline, family history, and neuroimaging findings.ResultsDocumentation rates for each phenotype varied in the structured versus unstructured EHR. Interannotator agreement was high (Cohen's kappa = 0.72-1) and positively correlated with the NLP-based phenotype extraction pipeline's performance (average F1-score = 0.65-0.99) for each phenotype.DiscussionWe developed an automated NLP-based pipeline to extract informative phenotypes that may improve the performance of eventual machine learning predictive models for AD. In the process, we examined documentation practices for each phenotype relevant to the care of AD patients and identified factors for success.ConclusionSuccess of our NLP-based phenotype extraction pipeline depended on domain-specific knowledge and focus on a specific clinical domain instead of maximizing generalizability.
| S-EPMC9952043 | biostudies-literature
Project description:BackgroundPolygenic risk scores (PRS) may enhance risk stratification for coronary heart disease among young adults. Whether a coronary heart disease PRS improves prediction beyond modifiable risk factors in this population is not known.MethodsGenotyped adults aged 18 to 35 years were selected from the CARDIA study (Coronary Artery Risk Development in Young Adults; n=1132) and FOS (Framingham Offspring Study; n=663). Systolic blood pressure, total and HDL (high-density lipoprotein) cholesterol, triglycerides, smoking, and waist circumference or body mass index were measured at the visit 1 exam of each study, and coronary artery calcium, a measure of coronary atherosclerosis, was assessed at year 15 (CARDIA) or year 30 (FOS). A previously validated PRS for coronary heart disease was computed for each subject. The C statistic and integrated discrimination improvement were used to compare improvements in prediction of elevated coronary artery calcium between models containing the PRS, risk factors, or both.ResultsThere were 62 (5%) and 93 (14%) participants with a coronary artery calcium score >20 (CARDIA) and >300 (FOS), respectively. At these thresholds, the C statistic changes of adding the PRS to a risk factor-based model were 0.015 (0.004-0.028) and 0.020 (0.001-0.039) in CARDIA and FOS, respectively. When adding risk factors to a PRS-based model, the respective changes were 0.070 (0.033-0.109) and 0.051 (0.017-0.079). The integrated discrimination improvement, when adding the PRS to a risk factor model, was 0.027 (-0.006 to 0.054) in CARDIA and 0.039 (0.0005-0.072) in FOS.ConclusionsAmong young adults, a PRS improved model discrimination for coronary atherosclerosis, but improvements were smaller than those associated with modifiable risk factors.
| S-EPMC8530876 | biostudies-literature
Project description:BackgroundThe electronic medical record contains a wealth of information buried in free text. We created a natural language processing algorithm to identify patients with atrial fibrillation (AF) using text alone.Methods and resultsWe created 3 data sets from patients with at least one AF billing code from 2010 to 2017: a training set (n=886), an internal validation set from site no. 1 (n=285), and an external validation set from site no. 2 (n=276). A team of clinicians reviewed and adjudicated patients as AF present or absent, which served as the reference standard. We trained 54 algorithms to classify each patient, varying the model, number of features, number of stop words, and the method used to create the feature set. The algorithm with the highest F-score (the harmonic mean of sensitivity and positive predictive value) in the training set was applied to the validation sets. F-scores and area under the receiver operating characteristic curves were compared between site no. 1 and site no. 2 using bootstrapping. Adjudicated AF prevalence was 75.1% at site no. 1 and 86.2% at site no. 2. Among 54 algorithms, the best performing model was logistic regression, using 1000 features, 100 stop words, and term frequency-inverse document frequency method to create the feature set, with sensitivity 92.8%, specificity 93.9%, and an area under the receiver operating characteristic curve of 0.93 in the training set. The performance at site no. 1 was sensitivity 92.5%, specificity 88.7%, with an area under the receiver operating characteristic curve of 0.91. The performance at site no. 2 was sensitivity 89.5%, specificity 71.1%, with an area under the receiver operating characteristic curve of 0.80. The F-score was lower at site no. 2 compared with site no. 1 (92.5% [SD, 1.1%] versus 94.2% [SD, 1.1%]; P<0.001).ConclusionsWe developed a natural language processing algorithm to identify patients with AF using text alone, with >90% F-score at 2 separate sites. This approach allows better use of the clinical narrative and creates an opportunity for precise, high-throughput cohort identification.
| S-EPMC7646941 | biostudies-literature
Project description:ObjectiveDelayed diagnosis of Kawasaki disease (KD) may lead to serious cardiac complications. We sought to create and test the performance of a natural language processing (NLP) tool, the KD-NLP, in the identification of emergency department (ED) patients for whom the diagnosis of KD should be considered.MethodsWe developed an NLP tool that recognizes the KD diagnostic criteria based on standard clinical terms and medical word usage using 22 pediatric ED notes augmented by Unified Medical Language System vocabulary. With high suspicion for KD defined as fever and three or more KD clinical signs, KD-NLP was applied to 253 ED notes from children ultimately diagnosed with either KD or another febrile illness. We evaluated KD-NLP performance against ED notes manually reviewed by clinicians and compared the results to a simple keyword search.ResultsKD-NLP identified high-suspicion patients with a sensitivity of 93.6% and specificity of 77.5% compared to notes manually reviewed by clinicians. The tool outperformed a simple keyword search (sensitivity = 41.0%; specificity = 76.3%).ConclusionsKD-NLP showed comparable performance to clinician manual chart review for identification of pediatric ED patients with a high suspicion for KD. This tool could be incorporated into the ED electronic health record system to alert providers to consider the diagnosis of KD. KD-NLP could serve as a model for decision support for other conditions in the ED.
| S-EPMC5031359 | biostudies-literature
Project description:Importance:Inpatient violence remains a significant problem despite existing risk assessment methods. The lack of robustness and the high degree of effort needed to use current methods might be mitigated by using routinely registered clinical notes. Objective:To develop and validate a multivariable prediction model for assessing inpatient violence risk based on machine learning techniques applied to clinical notes written in patients' electronic health records. Design, Setting, and Participants:This prognostic study used retrospective clinical notes registered in electronic health records during admission at 2 independent psychiatric health care institutions in the Netherlands. No exclusion criteria for individual patients were defined. At site 1, all adults admitted between January 2013 and August 2018 were included, and at site 2 all adults admitted to general psychiatric wards between June 2016 and August 2018 were included. Data were analyzed between September 2018 and February 2019. Main Outcomes and Measures:Predictive validity and generalizability of prognostic models measured using area under the curve (AUC). Results:Clinical notes recorded during a total of 3189 admissions of 2209 unique individuals at site 1 (mean [SD] age, 34.0 [16.6] years; 1536 [48.2%] male) and 3253 admissions of 1919 unique individuals at site 2 (mean [SD] age, 45.9 [16.6] years; 2097 [64.5%] male) were analyzed. Violent outcome was determined using the Staff Observation Aggression Scale-Revised. Nested cross-validation was used to train and evaluate models that assess violence risk during the first 4 weeks of admission based on clinical notes available after 24 hours. The predictive validity of models was measured at site 1 (AUC = 0.797; 95% CI, 0.771-0.822) and site 2 (AUC = 0.764; 95% CI, 0.732-0.797). The validation of pretrained models in the other site resulted in AUCs of 0.722 (95% CI, 0.690-0.753) at site 1 and 0.643 (95% CI, 0.610-0.675) at site 2; the difference in AUCs between the internally trained model and the model trained on other-site data was significant at site 1 (AUC difference = 0.075; 95% CI, 0.045-0.105; P < .001) and site 2 (AUC difference = 0.121; 95% CI, 0.085-0.156; P < .001). Conclusions and Relevance:Internally validated predictions resulted in AUC values with good predictive validity, suggesting that automatic violence risk assessment using routinely registered clinical notes is possible. The validation of trained models using data from other sites corroborates previous findings that violence risk assessment generalizes modestly to different populations.
| S-EPMC6613290 | biostudies-literature
Project description:Background and Aim: It was evaluated whether the integration of genetic risk scores (GRS-unweighted, wGRS-weighted) into conventional risk factor (CRF) models for coronary heart disease or acute myocardial infarction (CHD/AMI) could improve the predictive ability of the models. Methods: Subjects and data collected in a previous survey were used to perform regression and ROC curve analyses as well as to examine the role of genetic components. Thirty SNPs were selected, and genotype and phenotype data were available for 558 participants (general: N = 279 and Roma: N = 279). Results: The mean GRS (27.27 ± 3.43 vs. 26.68 ± 3.51, p = 0.046) and wGRS (3.52 ± 0.68 vs. 3.33 ± 0.62, p = 0.001) were significantly higher in the general population. The addition of the wGRS to the CRF model yielded the strongest improvement in discrimination among Roma (from 0.8616 to 0.8674), while the addition of GRS to the CRF model yielded the strongest improvement in discrimination in the general population (from 0.8149 to 0.8160). In addition to that, the Roma individuals were likely to develop CHD/AMI at a younger age than subjects in the general population. Conclusions: The combination of the CRFs and genetic components improved the model's performance and predicted AMI/CHD better than CRFs alone.
| S-EPMC10218435 | biostudies-literature
Project description:BackgroundCardiovascular disease (CVD) risk communication is a challenge for clinical practice, where physicians find it difficult to explain the absolute risk of a CVD event to patients with varying health literacy. Converting the probability to heart age is increasingly used to promote lifestyle change, but a rapid review of biological age interventions found no clear evidence that they motivate behavior change.ObjectiveIn this review, we aim to identify the content and effects of heart age interventions.MethodsWe conducted a systematic review of studies presenting heart age interventions to adults for CVD risk communication in April 2020 (later updated in March 2021). The Johanna Briggs risk of bias assessment tool was applied to randomized studies. Behavior change techniques described in the intervention methods were coded.ResultsFrom a total of 7926 results, 16 eligible studies were identified; these included 5 randomized web-based experiments, 5 randomized clinical trials, 2 mixed methods studies with quantitative outcomes, and 4 studies with qualitative analysis. Direct comparisons between heart age and absolute risk in the 5 web-based experiments, comprising 5514 consumers, found that heart age increased positive or negative emotional responses (4/5 studies), increased risk perception (4/5 studies; but not necessarily more accurate) and recall (4/4 studies), reduced credibility (2/3 studies), and generally had no effect on lifestyle intentions (4/5 studies). One study compared heart age and absolute risk to fitness age and found reduced lifestyle intentions for fitness age. Heart age combined with additional strategies (eg, in-person or phone counseling) in applied settings for 9582 patients improved risk control (eg, reduced cholesterol levels and absolute risk) compared with usual care in most trials (4/5 studies) up to 1 year. However, clinical outcomes were no different when directly compared with absolute risk (1/1 study). Mixed methods studies identified consultation time and content as important outcomes in actual consultations using heart age tools. There were differences between people receiving an older heart age result and those receiving a younger or equal to current heart age result. The heart age interventions included a wide range of behavior change techniques, and conclusions were sometimes biased in favor of heart age with insufficient supporting evidence. The risk of bias assessment indicated issues with all randomized clinical trials.ConclusionsThe findings of this review provide little evidence that heart age motivates lifestyle behavior change more than absolute risk, but either format can improve clinical outcomes when combined with other behavior change strategies. The label for the heart age concept can affect outcomes and should be pretested with the intended audience. Future research should consider consultation time and differentiate between results of older and younger heart age.International registered report identifier (irrid)NPRR2-10.1101/2020.05.03.20089938.
| S-EPMC8663444 | biostudies-literature
Project description:OBJECTIVE:In this work, we introduce a privacy technique for anonymizing clinical notes that guarantees all private health information is secured (including sensitive data, such as family history, that are not adequately covered by current techniques). MATERIALS AND METHODS:We employ a new "random replacement" paradigm (replacing each token in clinical notes with neighboring word vectors from the embedding space) to achieve 100% recall on the removal of sensitive information, unachievable with current "search-and-secure" paradigms. We demonstrate the utility of this paradigm on multiple corpora in a diverse set of classification tasks. RESULTS:We empirically evaluate the effect of our anonymization technique both on upstream and downstream natural language processing tasks to show that our perturbations, while increasing security (ie, achieving 100% recall on any dataset), do not greatly impact the results of end-to-end machine learning approaches. DISCUSSION:As long as current approaches utilize precision and recall to evaluate deidentification algorithms, there will remain a risk of overlooking sensitive information. Inspired by differential privacy, we sought to make it statistically infeasible to recreate the original data, although at the cost of readability. We hope that the work will serve as a catalyst to further research into alternative deidentification methods that can address current weaknesses. CONCLUSION:Our proposed technique can secure clinical texts at a low cost and extremely high recall with a readability trade-off while remaining useful for natural language processing classification tasks. We hope that our work can be used by risk-averse data holders to release clinical texts to researchers.
| S-EPMC7309261 | biostudies-literature
Project description:BackgroundSymptoms are a core concept of nursing interest. Large-scale secondary data reuse of notes in electronic health records (EHRs) has the potential to increase the quantity and quality of symptom research. However, the symptom language used in clinical notes is complex. A need exists for methods designed specifically to identify and study symptom information from EHR notes.ObjectivesWe aim to describe a method that combines standardized vocabularies, clinical expertise, and natural language processing to generate comprehensive symptom vocabularies and identify symptom information in EHR notes. We piloted this method with five diverse symptom concepts: constipation, depressed mood, disturbed sleep, fatigue, and palpitations.MethodsFirst, we obtained synonym lists for each pilot symptom concept from the Unified Medical Language System. Then, we used two large bodies of text (clinical notes from Columbia University Irving Medical Center and PubMed abstracts containing Medical Subject Headings or key words related to the pilot symptoms) to further expand our initial vocabulary of synonyms for each pilot symptom concept. We used NimbleMiner, an open-source natural language processing tool, to accomplish these tasks and evaluated NimbleMiner symptom identification performance by comparison to a manually annotated set of nurse- and physician-authored common EHR note types.ResultsCompared to the baseline Unified Medical Language System synonym lists, we identified up to 11 times more additional synonym words or expressions, including abbreviations, misspellings, and unique multiword combinations, for each symptom concept. Natural language processing system symptom identification performance was excellent.DiscussionUsing our comprehensive symptom vocabularies and NimbleMiner to label symptoms in clinical notes produced excellent performance metrics. The ability to extract symptom information from EHR notes in an accurate and scalable manner has the potential to greatly facilitate symptom science research.
| S-EPMC9109773 | biostudies-literature