Prediction of severe chest injury using natural language processing from the electronic health record.
Ontology highlight
ABSTRACT: Introduction
Trauma injury severity scores are currently calculated retrospectively from the electronic health record (EHR) using manual annotation by certified trauma coders. Natural language processing (NLP) of clinical documents in the EHR may enable automated injury scoring. We hypothesize that NLP with machine learning can discriminate between cases of severe and non-severe injury to the thorax after trauma.Methods
Clinical documents from a trauma center were examined between 2014 and 2018. Severe chest injury was defined as a thorax abbreviated injury score (AIS) >2 and served as the reference standard for supervised learning. Free text unigrams and concept unique identifiers (CUIs) from the Unified Medical Language Systems (UMLS) were extracted from clinical documents collected at one hour, four hours, and eight hours after patient arrival to the emergency department. Logistic regression models with elastic net regularization were tuned to maximize area under the receiver operating characteristic curve (AUROC) using 10-fold cross-validation on the training dataset (80%) and tested on a hold-out 20% dataset.Results
There were 6,891 traumas that met inclusion criteria. The complete data corpus consisted of 473,694 documents. Models trained using the first hour of data had a mean AUROC of 0.88 (95%CI [0.86, 0.89]); model discrimination and reclassification from the first hour significantly improved after eight hours with a mean AUROC of 0.94 (95%CI [0.93, 0.95]). Performance of models using CUIs were similar to unigrams (p>0.05). Models demonstrated excellent clinical face validity.Conclusions
Both CUIs and unigrams demonstrated excellent discrimination in predicting severity of chest injury using the first eight hours of clinical documents. Our model demonstrates that automated anatomical injury scoring is feasible and may be used for aggregation of data for trauma research and quality programs.
SUBMITTER: Kulshrestha S
PROVIDER: S-EPMC7856032 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:ImportanceAccurate prediction of outcomes among patients in intensive care units (ICUs) is important for clinical research and monitoring care quality. Most existing prediction models do not take full advantage of the electronic health record, using only the single worst value of laboratory tests and vital signs and largely ignoring information present in free-text notes. Whether capturing more of the available data and applying machine learning and natural language processing (NLP) can improve and automate the prediction of outcomes among patients in the ICU remains unknown.ObjectivesTo evaluate the change in power for a mortality prediction model among patients in the ICU achieved by incorporating measures of clinical trajectory together with NLP of clinical text and to assess the generalizability of this approach.Design, setting, and participantsThis retrospective cohort study included 101 196 patients with a first-time admission to the ICU and a length of stay of at least 4 hours. Twenty ICUs at 2 academic medical centers (University of California, San Francisco [UCSF], and Beth Israel Deaconess Medical Center [BIDMC], Boston, Massachusetts) and 1 community hospital (Mills-Peninsula Medical Center [MPMC], Burlingame, California) contributed data from January 1, 2001, through June 1, 2017. Data were analyzed from July 1, 2017, through August 1, 2018.Main outcomes and measuresIn-hospital mortality and model discrimination as assessed by the area under the receiver operating characteristic curve (AUC) and model calibration as assessed by the modified Hosmer-Lemeshow statistic.ResultsAmong 101 196 patients included in the analysis, 51.3% (n = 51 899) were male, with a mean (SD) age of 61.3 (17.1) years; their in-hospital mortality rate was 10.4% (n = 10 505). A baseline model using only the highest and lowest observed values for each laboratory test result or vital sign achieved a cross-validated AUC of 0.831 (95% CI, 0.830-0.832). In contrast, that model augmented with measures of clinical trajectory achieved an AUC of 0.899 (95% CI, 0.896-0.902; P < .001 for AUC difference). Further augmenting this model with NLP-derived terms associated with mortality further increased the AUC to 0.922 (95% CI, 0.916-0.924; P < .001). These NLP-derived terms were associated with improved model performance even when applied across sites (AUC difference for UCSF: 0.077 to 0.021; AUC difference for MPMC: 0.071 to 0.051; AUC difference for BIDMC: 0.035 to 0.043; P < .001) when augmenting with NLP at each site.Conclusions and relevanceIntensive care unit mortality prediction models incorporating measures of clinical trajectory and NLP-derived terms yielded excellent predictive performance and generalized well in this sample of hospitals. The role of these automated algorithms, particularly those using unstructured data from notes and other sources, in clinical research and quality improvement seems to merit additional investigation.
| S-EPMC6324323 | biostudies-literature
Project description:ObjectiveSocial determinants of health (SDOH) impact health outcomes and are documented in the electronic health record (EHR) through structured data and unstructured clinical notes. However, clinical notes often contain more comprehensive SDOH information, detailing aspects such as status, severity, and temporality. This work has two primary objectives: (1) develop a natural language processing information extraction model to capture detailed SDOH information and (2) evaluate the information gain achieved by applying the SDOH extractor to clinical narratives and combining the extracted representations with existing structured data.Materials and methodsWe developed a novel SDOH extractor using a deep learning entity and relation extraction architecture to characterize SDOH across various dimensions. In an EHR case study, we applied the SDOH extractor to a large clinical data set with 225 089 patients and 430 406 notes with social history sections and compared the extracted SDOH information with existing structured data.ResultsThe SDOH extractor achieved 0.86 F1 on a withheld test set. In the EHR case study, we found extracted SDOH information complements existing structured data with 32% of homeless patients, 19% of current tobacco users, and 10% of drug users only having these health risk factors documented in the clinical narrative.ConclusionsUtilizing EHR data to identify SDOH health risk factors and social needs may improve patient care and outcomes. Semantic representations of text-encoded SDOH information can augment existing structured data, and this more comprehensive SDOH representation can assist health systems in identifying and addressing these social needs.
| S-EPMC10354760 | biostudies-literature
Project description:OBJECTIVES:To demonstrate the utility of a natural language processing (NLP) algorithm for mining kidney stone composition in a large-scale electronic health records (EHR) repository. METHODS:We developed StoneX, a pattern-matching method for extracting kidney stone composition information from clinical notes. We trained the extraction algorithm on manually annotated text mentions of calcium oxalate monohydrate, calcium oxalate dihydrate, hydroxyapatite, brushite, uric acid, and struvite stones. We employed StoneX to identify patients with kidney stone composition data and mine >125 million notes from our institutional EHR. Analyses performed on the extracted patients included stone type conversions over time, survival analysis from a second stone surgery, and disease associations by stone composition to validate the phenotyping method against known associations. RESULTS:The NLP algorithm identified 45,235 text mentions corresponding to 11,585 patients. Overall, the system achieved positive predictive value >90% for calcium oxalate monohydrate, calcium oxalate dihydrate, hydroxyapatite, brushite, and struvite; except for uric acid (positive predictive value?=?87.5%). Survival analysis from a second stone surgery showed statistically significant differences among stone types (P?=?.03). Several phenotype associations were found: uric acid-type 2 diabetes (odds ratio, OR?=?2.69, 95% confidence intervals, CI?=?1.91-3.79), struvite-neurogenic bladder (OR?=?12.27, 95% CI?=?4.33-34.79), struvite-urinary tract infection (OR?=?7.36, 95% CI?=?3.01-17.99), hydroxyapatite-pulmonary collapse (OR?=?3.67, 95% CI?=?2.10-6.42), hydroxyapatite-neurogenic bladder (OR?=?5.23, 95% CI?=?2.05-13.36), brushite-calcium metabolism disorder (OR?=?4.59, 95% CI?=?2.14-9.81), and brushite-hypercalcemia (OR?=?4.09, 95% CI?=?1.90-8.80). CONCLUSION:NLP extraction of kidney stone composition from large-scale EHRs is feasible with high precision, enabling high-throughput epidemiological studies of kidney stone disease. These tools will enable high fidelity kidney stone research from the EHR.
| S-EPMC6778032 | biostudies-literature
Project description:ContextGoals-of-care discussions are an important quality metric in palliative care. However, goals-of-care discussions are often documented as free text in diverse locations. It is difficult to identify these discussions in the electronic health record (EHR) efficiently.ObjectivesTo develop, train, and test an automated approach to identifying goals-of-care discussions in the EHR, using natural language processing (NLP) and machine learning (ML).MethodsFrom the electronic health records of an academic health system, we collected a purposive sample of 3183 EHR notes (1435 inpatient notes and 1748 outpatient notes) from 1426 patients with serious illness over 2008-2016, and manually reviewed each note for documentation of goals-of-care discussions. Separately, we developed a program to identify notes containing documentation of goals-of-care discussions using NLP and supervised ML. We estimated the performance characteristics of the NLP/ML program across 100 pairs of randomly partitioned training and test sets. We repeated these methods for inpatient-only and outpatient-only subsets.ResultsOf 3183 notes, 689 contained documentation of goals-of-care discussions. The mean sensitivity of the NLP/ML program was 82.3% (SD 3.2%), and the mean specificity was 97.4% (SD 0.7%). NLP/ML results had a median positive likelihood ratio of 32.2 (IQR 27.5-39.2) and a median negative likelihood ratio of 0.18 (IQR 0.16-0.20). Performance was better in inpatient-only samples than outpatient-only samples.ConclusionUsing NLP and ML techniques, we developed a novel approach to identifying goals-of-care discussions in the EHR. NLP and ML represent a potential approach toward measuring goals-of-care discussions as a research outcome and quality metric.
| S-EPMC7769906 | biostudies-literature
Project description:BackgroundUsing novel data mining methods such as natural language processing (NLP) on electronic health records (EHRs) for screening and detecting individuals at risk for psychosis.MethodThe study included all patients receiving a first index diagnosis of nonorganic and nonpsychotic mental disorder within the South London and Maudsley (SLaM) NHS Foundation Trust between January 1, 2008, and July 28, 2018. Least Absolute Shrinkage and Selection Operator (LASSO)-regularized Cox regression was used to refine and externally validate a refined version of a five-item individualized, transdiagnostic, clinically based risk calculator previously developed (Harrell's C = 0.79) and piloted for implementation. The refined version included 14 additional NLP-predictors: tearfulness, poor appetite, weight loss, insomnia, cannabis, cocaine, guilt, irritability, delusions, hopelessness, disturbed sleep, poor insight, agitation, and paranoia.ResultsA total of 92 151 patients with a first index diagnosis of nonorganic and nonpsychotic mental disorder within the SLaM Trust were included in the derivation (n = 28 297) or external validation (n = 63 854) data sets. Mean age was 33.6 years, 50.7% were women, and 67.0% were of white race/ethnicity. Mean follow-up was 1590 days. The overall 6-year risk of psychosis in secondary mental health care was 3.4 (95% CI, 3.3-3.6). External validation indicated strong performance on unseen data (Harrell's C 0.85, 95% CI 0.84-0.86), an increase of 0.06 from the original model.ConclusionsUsing NLP on EHRs can considerably enhance the prognostic accuracy of psychosis risk calculators. This can help identify patients at risk of psychosis who require assessment and specialized care, facilitating earlier detection and potentially improving patient outcomes.
| S-EPMC7965059 | biostudies-literature
Project description:BackgroundAbstracting cerebrovascular disease (CeVD) from inpatient electronic medical records (EMRs) through natural language processing (NLP) is pivotal for automated disease surveillance and improving patient outcomes. Existing methods rely on coders' abstraction, which has time delays and under-coding issues. This study sought to develop an NLP-based method to detect CeVD using EMR clinical notes.MethodsCeVD status was confirmed through a chart review on randomly selected hospitalized patients who were 18 years or older and discharged from 3 hospitals in Calgary, Alberta, Canada, between January 1 and June 30, 2015. These patients' chart data were linked to administrative discharge abstract database (DAD) and Sunrise™ Clinical Manager (SCM) EMR database records by Personal Health Number (a unique lifetime identifier) and admission date. We trained multiple natural language processing (NLP) predictive models by combining two clinical concept extraction methods and two supervised machine learning (ML) methods: random forest and XGBoost. Using chart review as the reference standard, we compared the model performances with those of the commonly applied International Classification of Diseases (ICD-10-CA) codes, on the metrics of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).ResultOf the study sample (n = 3036), the prevalence of CeVD was 11.8% (n = 360); the median patient age was 63; and females accounted for 50.3% (n = 1528) based on chart data. Among 49 extracted clinical documents from the EMR, four document types were identified as the most influential text sources for identifying CeVD disease ("nursing transfer report," "discharge summary," "nursing notes," and "inpatient consultation."). The best performing NLP model was XGBoost, combining the Unified Medical Language System concepts extracted by cTAKES (e.g., top-ranked concepts, "Cerebrovascular accident" and "Transient ischemic attack"), and the term frequency-inverse document frequency vectorizer. Compared with ICD codes, the model achieved higher validity overall, such as sensitivity (25.0% vs 70.0%), specificity (99.3% vs 99.1%), PPV (82.6 vs. 87.8%), and NPV (90.8% vs 97.1%).ConclusionThe NLP algorithm developed in this study performed better than the ICD code algorithm in detecting CeVD. The NLP models could result in an automated EMR tool for identifying CeVD cases and be applied for future studies such as surveillance, and longitudinal studies.
| S-EPMC10474977 | biostudies-literature
Project description:ImportanceInternational Classification of Diseases-coded hospital discharge data do not accurately reflect whether firearm injuries were caused by assault, unintentional injury, self-harm, legal intervention, or were of undetermined intent. Applying natural language processing (NLP) and machine learning (ML) techniques to electronic health record (EHR) narrative text could be associated with improved accuracy of firearm injury intent data.ObjectiveTo assess the accuracy with which an ML model identified firearm injury intent.Design, setting, and participantsA cross-sectional retrospective EHR review was conducted at 3 level I trauma centers, 2 from health care institutions in Boston, Massachusetts, and 1 from Seattle, Washington, between January 1, 2000, and December 31, 2019; data analysis was performed from January 18, 2021, to August 22, 2022. A total of 1915 incident cases of firearm injury in patients presenting to emergency departments at the model development institution and 769 from the external validation institution with a firearm injury code assigned according to International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) or International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Clinical Modification (ICD-10-CM), in discharge data were included.ExposuresClassification of firearm injury intent.Main outcomes and measuresIntent classification accuracy by the NLP model was compared with ICD codes assigned by medical record coders in discharge data. The NLP model extracted intent-relevant features from narrative text that were then used by a gradient-boosting classifier to determine the intent of each firearm injury. Classification accuracy was evaluated against intent assigned by the research team. The model was further validated using an external data set.ResultsThe NLP model was evaluated in 381 patients presenting with firearm injury at the model development site (mean [SD] age, 39.2 [13.0] years; 348 [91.3%] men) and 304 patients at the external development site (mean [SD] age, 31.8 [14.8] years; 263 [86.5%] men). The model proved more accurate than medical record coders in assigning intent to firearm injuries at the model development site (accident F-score, 0.78 vs 0.40; assault F-score, 0.90 vs 0.78). The model maintained this improvement on an external validation set from a second institution (accident F-score, 0.64 vs 0.58; assault F-score, 0.88 vs 0.81). While the model showed some degradation between institutions, retraining the model using data from the second institution further improved performance on that site's records (accident F-score, 0.75; assault F-score, 0.92).Conclusions and relevanceThe findings of this study suggest that NLP ML can be used to improve the accuracy of firearm injury intent classification compared with ICD-coded discharge data, particularly for cases of accident and assault intents (the most prevalent and commonly misclassified intent types). Future research could refine this model using larger and more diverse data sets.
| S-EPMC10080369 | biostudies-literature
Project description:ObjectiveLimited research exists in predicting first-time suicide attempts that account for two-thirds of suicide decedents. We aimed to predict first-time suicide attempts using a large data-driven approach that applies natural language processing (NLP) and machine learning (ML) to unstructured (narrative) clinical notes and structured electronic health record (EHR) data.MethodsThis case-control study included patients aged 10-75 years who were seen between 2007 and 2016 from emergency departments and inpatient units. Cases were first-time suicide attempts from coded diagnosis; controls were randomly selected without suicide attempts regardless of demographics, following a ratio of nine controls per case. Four data-driven ML models were evaluated using 2-year historical EHR data prior to suicide attempt or control index visits, with prediction windows from 7 to 730 days. Patients without any historical notes were excluded. Model evaluation on accuracy and robustness was performed on a blind dataset (30% cohort).ResultsThe study cohort included 45 238 patients (5099 cases, 40 139 controls) comprising 54 651 variables from 5.7 million structured records and 798 665 notes. Using both unstructured and structured data resulted in significantly greater accuracy compared to structured data alone (area-under-the-curve [AUC]: 0.932 vs. 0.901 P < .001). The best-predicting model utilized 1726 variables with AUC = 0.932 (95% CI, 0.922-0.941). The model was robust across multiple prediction windows and subgroups by demographics, points of historical most recent clinical contact, and depression diagnosis history.ConclusionsOur large data-driven approach using both structured and unstructured EHR data demonstrated accurate and robust first-time suicide attempt prediction, and has the potential to be deployed across various populations and clinical settings.
| S-EPMC7966858 | biostudies-literature
Project description:BackgroundThe prognosis, diagnosis, and treatment of many genetic disorders and familial diseases significantly improve if the family history (FH) of a patient is known. Such information is often written in the free text of clinical notes.ObjectiveThe aim of this study is to develop automated methods that enable access to FH data through natural language processing.MethodsWe performed information extraction by using transformers to extract disease mentions from notes. We also experimented with rule-based methods for extracting family member (FM) information from text and coreference resolution techniques. We evaluated different transfer learning strategies to improve the annotation of diseases. We provided a thorough error analysis of the contributing factors that affect such information extraction systems.ResultsOur experiments showed that the combination of domain-adaptive pretraining and intermediate-task pretraining achieved an F1 score of 81.63% for the extraction of diseases and FMs from notes when it was tested on a public shared task data set from the National Natural Language Processing Clinical Challenges (N2C2), providing a statistically significant improvement over the baseline (P<.001). In comparison, in the 2019 N2C2/Open Health Natural Language Processing Shared Task, the median F1 score of all 17 participating teams was 76.59%.ConclusionsOur approach, which leverages a state-of-the-art named entity recognition model for disease mention detection coupled with a hybrid method for FM mention detection, achieved an effectiveness that was close to that of the top 3 systems participating in the 2019 N2C2 FH extraction challenge, with only the top system convincingly outperforming our approach in terms of precision.
| S-EPMC8092929 | biostudies-literature
Project description:ObjectiveTo address thyroid cancer overdiagnosis, we aim to develop a natural language processing (NLP) algorithm to determine the appropriateness of thyroid ultrasounds (TUS).Patients and methodsBetween 2017 and 2021, we identified 18,000 TUS patients at Mayo Clinic and selected 628 for chart review to create a ground truth dataset based on consensus. We developed a rule-based NLP pipeline to identify TUS as appropriate TUS (aTUS) or inappropriate TUS (iTUS) using patients' clinical notes and additional meta information. In addition, we designed an abbreviated NLP pipeline (aNLP) solely focusing on labels from TUS order requisitions to facilitate deployment at other health care systems. Our dataset was split into a training set of 468 (75%) and a test set of 160 (25%), using the former for rule development and the latter for performance evaluation.ResultsThere were 449 (95.9%) patients identified as aTUS and 19 (4.06%) as iTUS in the training set; there are 155 (96.88%) patients identified as aTUS and 5 (3.12%) were iTUS in the test set. In the training set, the pipeline achieved a sensitivity of 0.99, specificity of 0.95, and positive predictive value of 1.0 for detecting aTUS. The testing cohort revealed a sensitivity of 0.96, specificity of 0.80, and positive predictive value of 0.99. Similar performance metrics were observed in the aNLP pipeline.ConclusionThe NLP models can accurately identify the appropriateness of a thyroid ultrasound from clinical documentation and order requisition information, a critical initial step toward evaluating the drivers and outcomes of TUS use and subsequent thyroid cancer overdiagnosis.
| S-EPMC10947349 | biostudies-literature