Rule-based Natural Language Processing Research Articles

Connecting the use of innovative treatments and glucocorticoids with the multidisciplinary evaluation through rule-based natural-language processing: a real-world study on patients with rheumatoid arthritis, psoriatic arthritis, and psoriasis.

The impact of a multidisciplinary management of rheumatoid arthritis (RA), psoriatic arthritis (PsA), and psoriasis on systemic glucocorticoids or innovative treatments remains unknown. Rule-based natural language processing and text extraction help to manage large datasets of unstructured information and provide insights into the profile of treatment choices. We obtained structured information from text data of outpatient visits between 2017 and 2022 using regular expressions (RegEx) to define elastic search patterns and to consider only affirmative citation of diseases or prescribed therapy by detecting negations. Care processes were described by binary flags which express the presence of RA, PsA and psoriasis and the prescription of glucocorticoids and biologics or small molecules in each cases. Logistic regression analyses were used to train the classifier to predict outcomes using the number of visits and the other specialist visits as the main variables. We identified 1743 patients with RA, 1359 with PsA and 2,287 with psoriasis, accounting for 5,677, 4,468 and 7,770 outpatient visits, respectively. Among these, 25% of RA, 32% of PsA and 25% of psoriasis cases received biologics or small molecules, while 49% of RA, 28% of PsA, and 40% of psoriasis cases received glucocorticoids. Patients evaluated also by other specialists were treated more frequently with glucocorticoids (70% vs. 49% for RA, 60% vs. 28% for PsA, 51% vs. 40% for psoriasis; p < 0.001) as well as with biologics/small molecules (49% vs. 25% for RA, 64% vs. 32% in PsA; 51% vs. 25% for psoriasis; p < 0.001) compared to cases seen only by the main specialist. Patients with RA, PsA, or psoriasis undergoing multiple evaluations are more likely to receive innovative treatments or glucocorticoids, possibly reflecting more complex cases.

Analyzing free-text clinical narratives for veterans with lymphoid malignancies using natural language processing (NLP).

e13576 Background: Free-text clinical narratives contain rich patient information, which is labor-intensive to extract through chart review. We developed an NLP pipeline to enable automatic extraction of performance status (PS), staging, and diagnosis from clinical narratives from Veterans Affairs (VA) patients with lymphoid malignancies (LM). Methods: The rule-based NLP algorithm was developed and iteratively refined using a development corpus of 287 notes independently annotated by two clinicians. The F1-score for PS was 95.8 (precision 98.6, recall 93.2), 92.7 for staging (precision 94.0, recall 81.6), and 67 (precision 80.2, recall 57.9) for diagnosis. The NLP pipeline was then externally validated using an evaluation corpus of 97 notes from another group of 100 veterans with T-cell LM. Results: The results are reported. In the 97 notes, primary diagnosis was most routinely documented, with 2.76 mentions per note. In comparison, staging was most sparsely documented with only 34 mentions (note that 11 patients with large granular lymphocytic leukemia were not staged). The NLP pipeline performed relatively well in extracting PS and staging (F1-scores were 0.74 and 0.72, respectively). It also achieved high precision in extracting diagnosis information (precision 0.93). However, recall (0.44) for diagnosis was poor, likely due to the complexities and inconsistencies of how diagnoses are documented for LM. Frequency of documentation and performance in the external validation set. Conclusions: The pipeline shows promising performance on the external validation set, demonstrating the feasibility of using NLP to extract information from notes of patients with LM for clinical research. The NLP pipeline generally has lower recall than precision, indicating that the pipeline may miss clinical information that should be captured. FPs incorrectly capture entities that are easily confused with the clinical entities of interest, such as nutritional status versus performance status. Future work includes capturing more lexical variations and indicators of documentation, as well as contextual information, such as in which sections of notes elements are likely documented. In addition, we describe how diagnosis may exist as primary, secondary, and in a differential, and we are building an NLP-based classifier to distinguish between these types of diagnoses. We will use results from the rule-based NLP pipeline as labels to fine-tune transformer-based, weak-supervised models to further enhance the performance. [Table: see text]

Application of natural language processing to assess the performance status documentation quality metric in patients with non–small-cell lung cancer.

e13582 Background: Performance status (PS) assessment is important for cancer prognosis and treatment decision-making. PS documentation for stage IV non-small cell lung cancer (NSCLC) is a quality measure for ASCO Quality Oncology Practice Initiative (QOPI) Certification Program. We used natural language processing (NLP) to extract PS from unstructured electronic medical records (EMR) clinical notes in patients with metastatic NSCLC to verify PS documentation compliance and disparities in PS documentation. Methods: We identified oncology progress notes of stage IV NSCLC patients from Harris Health System, which is an integrated comprehensive healthcare system for underserved patients in Houston, TX. All progress notes within 60 days of first oncology clinic visit date were filtered based on visit type and author type. A rule-based regular expression NLP algorithm was used to extract the PS, namely Eastern Standard Cooperative Group (ECOG) or Karnofsky Performance Status (KPS). A total of 400 oncology notes were randomly selected from 400 distinct patients and annotated by two clinician reviewers, and discrepancies were resolved by a senior clinician judge. The annotated notes were split randomly into 200 training notes and 200 validation notes, and the performance of the NLP algorithm was evaluated. We used logistic regression to analyze potential disparity in PS documentation. Results: We identified 15,239 progress notes from 757 stage IV NSCLC patients. The NLP algorithm accuracy validated on the test data was 99% with corresponding positive predictive values (PPVs) for ECOG PS 0, 1, 2, 3, 4 of 100%, 100%, 95%, 94%, 94%, respectively. The NLP algorithm identified 85% of patients with appropriate PS documentation (sensitivity 100%, specificity 91%, positive predictive value 98%, negative predictive value 100%). Logistic regression analyses showed no significant differences in association between PS documentation and race/ethnicity, language, or sex. Conclusions: We developed an internally validated NLP algorithm to extract PS from EMR clinical notes. Despite potential language barriers, the QOPI metric of &gt;75% PS documentation was met in our hospital system without significant disparities in clinical care. [Table: see text]

Investigating online activity in UK adolescent mental health patients: a feasibility study using a natural language processing approach for electronic health records

ObjectivesTo assess the feasibility of using a natural language processing (NLP) application for extraction of free-text online activity mentions in adolescent mental health patient electronic health records (EHRs).SettingThe Clinical Records...

Multistep Automated Data Labelling Procedure (MADLaP) for thyroid nodules on ultrasound: An artificial intelligence approach for automating image annotation

Machine learning (ML) for diagnosis of thyroid nodules on ultrasound is an active area of research. However, ML tools require large, well-labeled datasets, the curation of which is time-consuming and labor-intensive. The purpose of our study was to develop and test a deep-learning-based tool to facilitate and automate the data annotation process for thyroid nodules; we named our tool Multistep Automated Data Labelling Procedure (MADLaP). MADLaP was designed to take multiple inputs including pathology reports, ultrasound images, and radiology reports. Using multiple step-wise ‘modules’ including rule-based natural language processing, deep-learning-based imaging segmentation, and optical character recognition, MADLaP automatically identified images of a specific thyroid nodule and correctly assigned a pathology label. The model was developed using a training set of 378 patients across our health system and tested on a separate set of 93 patients. Ground truths for both sets were selected by an experienced radiologist. Performance metrics including yield (how many labeled images the model produced) and accuracy (percentage correct) were measured using the test set. MADLaP achieved a yield of 63 % and an accuracy of 83 %. The yield progressively increased as the input data moved through each module, while accuracy peaked part way through. Error analysis showed that inputs from certain examination sites had lower accuracy (40 %) than the other sites (90 %, 100 %). MADLaP successfully created curated datasets of labeled ultrasound images of thyroid nodules. While accurate, the relatively suboptimal yield of MADLaP exposed some challenges when trying to automatically label radiology images from heterogeneous sources. The complex task of image curation and annotation could be automated, allowing for enrichment of larger datasets for use in machine learning development.

Near Real-time Natural Language Processing for the Extraction of Abdominal Aortic Aneurysm Diagnoses From Radiology Reports: Algorithm Development and Validation Study.

Management of abdominal aortic aneurysms (AAAs) requires serial imaging surveillance to evaluate the aneurysm dimension. Natural language processing (NLP) has been previously developed to retrospectively identify patients with AAA from electronic health records (EHRs). However, there are no reported studies that use NLP to identify patients with AAA in near real-time from radiology reports. This study aims to develop and validate a rule-based NLP algorithm for near real-time automatic extraction of AAA diagnosis from radiology reports for case identification. The AAA-NLP algorithm was developed and deployed to an EHR big data infrastructure for near real-time processing of radiology reports from May 1, 2019, to September 2020. NLP extracted named entities for AAA case identification and classified subjects as cases and controls. The reference standard to assess algorithm performance was a manual review of processed radiology reports by trained physicians following standardized criteria. Reviewers were blinded to the diagnosis of each subject. The AAA-NLP algorithm was refined in 3 successive iterations. For each iteration, the AAA-NLP algorithm was modified based on performance compared to the reference standard. A total of 360 reports were reviewed, of which 120 radiology reports were randomly selected for each iteration. At each iteration, the AAA-NLP algorithm performance improved. The algorithm identified AAA cases in near real-time with high positive predictive value (0.98), sensitivity (0.95), specificity (0.98), F1 score (0.97), and accuracy (0.97). Implementation of NLP for accurate identification of AAA cases from radiology reports with high performance in near real time is feasible. This NLP technique will support automated input for patient care and clinical decision support tools for the management of patients with AAA. .

Novel use of natural language processing for registry development in peritoneal surface malignancies

BackgroundTraditional methods of research registry development for rare conditions such as peritoneal surface malignancies (PSM) are often hindered by poor patient accrual and need for significant manpower resources. We develop a novel pipeline using natural language processing (NLP) to accelerate this process and demonstrate its real-world application in the identification of PSM patients, as well as characterisation of referral patterns in this cohort. Materials and methodsA training set comprising 100 radiological reports of abdomen and pelvis computed tomography scans was used to develop a rule-based NLP system able to classify reports based on the presence or absence of PSM. The algorithm was applied to a test set of 10,261 reports to identify all patients with PSM for registry creation. The registry was subsequently linked to electronic medical records, and the referral patterns of patients evaluated. ResultsThe algorithm identified 251 reports as positive for PSM from a total of 10,261 reports, of which 239 were concordant with manual review. Performance was excellent with a specificity of 90%, positive predictive value of 95%, and accuracy of 96%. From these, 228 unique patients were identified for registry inclusion after corroboration with pathological findings. Only 27.6% of patients were found to have been referred to and reviewed by PSM specialist surgeons. For those without a PSM specialist consult, 39.4% were managed by medical oncology, 11.5% by colorectal surgery, 7.3% by gastroenterology, 5.4% by internal medicine, and 29.1% by various other miscellaneous medical and surgical subspecialties. ConclusionNLP is a useful tool in automated pipelines that can greatly contribute to registry creation, as well as research and quality improvement efforts.

Automatic Classification of Tumor Response From Radiology Reports With Rule-Based Natural Language Processing Integrated Into the Clinical Oncology Workflow.

Imaging reports in oncology provide critical information about the disease evolution that should be timely shared to tailor the clinical decision making and care coordination of patients with advanced cancer. However, tumor response stays unstructured in free-text and underexploited. Natural language processing (NLP) methods can help provide this critical information into the electronic health records (EHR) in real time to assist health care workers. A rule-based algorithm was developed using SAS tools to automatically extract and categorize tumor response within progression or no progression categories. 2,970 magnetic resonance imaging, computed tomography scan, and positron emission tomography French reports were extracted from the EHR of a large comprehensive cancer center to build a 2,637-document training set and a 603-document validation set. The model was also tested on 189 imaging reports from 46 different radiology centers. A tumor dashboard was created in the EHR using the Timeline tool of the vis.js javascript library. An NLP methodology was applied to create an ontology of radiographic terms defining tumor response, mapping text to five main concepts, and application decision rules on the basis of clinical practice RECIST guidelines. The model achieved an overall accuracy of 0.88 (ranging from 0.87 to 0.94), with similar performance on both progression and no progression classification. The overall accuracy was 0.82 on reports from different radiology centers. Data were visualized and organized in a dynamic tumor response timeline. This tool was deployed successfully at our institution both retrospectively and prospectively as part of an automatic pipeline to screen reports and classify tumor response in real time for all metastatic patients. Our approach provides an NLP-based framework to structure and classify tumor response from the EHR and integrate tumor response classification into the clinical oncology workflow.

Assess the documentation of cognitive tests and biomarkers in electronic health records via natural language processing for Alzheimer’s disease and related dementias

BackgroundCognitive tests and biomarkers are the key information to assess the severity and track the progression of Alzheimer’s’ disease (AD) and AD-related dementias (AD/ADRD), yet, both are often only documented in clinical narratives of patients’ electronic health records (EHRs). In this work, we aim to (1) assess the documentation of cognitive tests and biomarkers in EHRs that can be used as real-world endpoints, and (2) identify, extract, and harmonize the different commonly used cognitive tests from clinical narratives using natural language processing (NLP) methods into categorical AD/ADRD severity. MethodsWe developed a rule-based NLP pipeline to extract the cognitive tests and biomarkers from clinical narratives in AD/ADRD patients’ EHRs. We aggregated the extracted results to the patient level and harmonized the cognitive test scores into severity categories using cutoffs determined based on both relevant literature and domain knowledge of AD/ADRD clinicians. ResultsWe identified an AD/ADRD cohort of 48,912 patients from the University of Florida (UF) Health system and identified 7 measurements (6 cognitive tests and 1 biomarker) that are frequently documented in our data. Our NLP pipeline achieved an overall F1-score of 0.9059 across the 7 measurements. Among the 6 cognitive tests, we were able to harmonize 4 cognitive test scores into severity categories, and the population characteristics of patients with different severity were described. We also identified several factors related to the availability of their documentation in EHRs. ConclusionThis study demonstrates that our NLP pipelines can extract cognitive tests and biomarkers of AD/ADRD accurately for downstream studies. Although, the documentation of cognitive tests and biomarkers in EHRs appears to be low, RWD is still an important resource for AD/ADRD research. Nevertheless, providing standardized approach to document cognitive tests and biomarkers in EHRS are also warranted.

LEVERAGING NATURAL LANGUAGE PROCESSING TO IDENTIFY CAREGIVER AVAILABILITY FOR PATIENTS WITH ALZHEIMER'S DISEASE

Abstract Identifying caregiver availability for patients with Alzheimer’s disease and related dementia (ADRD) can inform healthcare systems and hospital administrations on appropriate discharge planning. However, caregiver availability data is not systematically collected in the healthcare system. Our objective was to use medical notes (from physicians, nurses, or social workers) to assess caregiver availability and types of caregivers for hospitalized patients with ADRD. We used 2016-2019 telephone-encounter notes from a single institution. Rule-based natural language processing (NLP) was used to develop an algorithm to identify whether the patient (1) resides at home; (2) resides at an institution; (3) has a formal caregiver; and (4) has an informal caregiver. We evaluated the data at the note level and used chart abstraction as the “gold standard.” We examined the validity of the algorithm for both training (n=749) and test sets (n=227). Our results indicated a high level of accuracy and reliability for identifying an informal caregiver (F1=0.942; accuracy=0.947, sensitivity=0.970, and specificity=0.928). Identifying whether the patient lived at an institution was the least reliable measure (F1=0.638, accuracy=0.899, sensitivity=0.512, specificity=0.978). The most common causes of misclassifications were: (1) incomplete or misspelled names of the facilities; (2) past/uncertain/undecided situations; (3) lack of specificity; (4) use of uncommon abbreviations; and (5) irregular use of templates. Our NLP algorithm was able to identify whether the patient lived at home vs. an institution and whether the patient had a formal or informal caregiver. There is merit in continuing the NLP approach to identify more granular, caregiver-related information (e.g., hours available).

Using natural language processing to identify opioid use disorder in electronic health record data

BackgroundAs opioid prescriptions have risen, there has also been an increase in opioid use disorder (OUD) and its adverse outcomes. Accurate and complete epidemiologic surveillance of OUD, to inform prevention strategies, presents challenges. The objective of this study was to ascertain prevalence of OUD using two methods to identify OUD in electronic health records (EHR): applying natural language processing (NLP) for text mining of unstructured clinical notes and using ICD-10-CM diagnostic codes. MethodsData were drawn from EHR records for hospital and emergency department patient visits to a large regional academic medical center from 2017 to 2019. International Classification of Disease, 10th Edition, Clinic Modification (ICD-10-CM) discharge codes were extracted for each visit. To develop the rule-based NLP algorithm, a stepwise process was used. First, a small sample of visits from 2017 was used to develop initial dictionaries. Next, EHR corresponding to 30,124 visits from 2018 were used to develop and evaluate the rule-based algorithm. A random sample of the results were manually reviewed to identify and address shortcomings in the algorithm, and to estimate sensitivity and specificity of the two methods of ascertainment. Last, the final algorithm was then applied to 29,212 visits from 2019 to estimate OUD prevalence. ResultsWhile there was substantial overlap in the identified records (n = 1,381 [59.2 %]), overall n = 2,332 unique visits were identified. Of the total unique visits, 430 (18.4 %) were identified only by ICD-10-CM codes, and 521 (22.3 %) were identified only by NLP. The prevalence of visits with evidence of an OUD diagnosis in this sample, ascertained using only ICD-10-CM codes, was 1,811/29,212 (6.1 %). Including the additional 521 visits identified only by NLP, the estimated prevalence of OUD is 2,332/29,212 (7.9 %), an increase of 29.5 % compared to the use of ICD-10-CM codes alone. The estimated sensitivity and specificity of the NLP-based OUD classification were 81.8 % and 97.5 %, respectively, relative to gold-standard manual review by an expert addiction medicine physician. ConclusionNLP-based algorithms can automate data extraction and identify evidence of opioid use disorder from unstructured electronic healthcare records. The most complete ascertainment of OUD in EHR was combined NLP with ICD-10-CM codes. NLP should be considered for epidemiological studies involving EHR data.

Using Primary Care Clinical Text Data and Natural Language Processing to Identify Indicators of COVID-19 in Toronto, Canada.

The objective of this study was to investigate whether a rule-based natural language processing (NLP) system, applied to primary care clinical text data, could be used to monitor COVID-19 viral activity in Toronto, Canada. We employed a retrospective cohort design. We included primary care patients with a clinical encounter between January 1, 2020 and December 31, 2020 at one of 44 participating clinical sites. During the study timeframe, Toronto first experienced a COVID-19 outbreak between March-2020 and June-2020; followed by a second viral resurgence from October-2020 through December-2020. We used an expert derived dictionary, pattern matching tools and contextual analyzer to classify primary care documents as 1) COVID-19 positive, 2) COVID-19 negative, or 3) unknown COVID-19 status. We applied the COVID-19 biosurveillance system across three primary care electronic medical record text streams: 1) lab text, 2) health condition diagnosis text and 3) clinical notes. We enumerated COVID-19 entities in the clinical text and estimated the proportion of patients with a positive COVID-19 record. We constructed a primary care COVID-19 NLP-derived time series and investigated its correlation with independent/external public health series: 1) lab confirmed COVID-19 cases, 2) COVID-19 hospitalizations, 3) COVID-19 ICU admissions, and 4) COVID-19 intubations. A total of 196,440 unique patients were observed over the study timeframe, of which 4,580 (2.3%) had at least one positive COVID-19 document in their primary care electronic medical record. Our NLP-derived COVID-19 time series describing the temporal dynamics of COVID-19 positivity status over the study timeframe demonstrated a pattern/trend which strongly mirrored that of other external public health series under investigation. We conclude that primary care text data passively collected from electronic medical record systems represent a high quality, low-cost source of information for monitoring/surveilling COVID-19 impacts on community health.

Using Natural Language Processing to Identify Low Back Pain in Imaging Reports

A natural language processing (NLP) pipeline was developed to identify lumbar spine imaging findings associated with low back pain (LBP) in X-radiation (X-ray), computed tomography (CT), and magnetic resonance imaging (MRI) reports. A total of 18,640 report datasets were randomly sampled (stratified by imaging modality) to obtain a balanced sample of 300 X-ray, 300 CT, and 300 MRI reports. A total of 23 radiologic findings potentially related to LBP were defined, and their presence was extracted from radiologic reports. In developing NLP pipelines, section and sentence segmentation from the radiology reports was performed using a rule-based method, including regular expression with negation detection. Datasets were randomly split into 80% for development and 20% for testing to evaluate the model’s extraction performance. The performance of the NLP pipeline was evaluated by using recall, precision, accuracy, and the F1 score. In evaluating NLP model performances, four parameters—recall, precision, accuracy, and F1 score—were greater than 0.9 for all 23 radiologic findings. These four scores were 1.0 for 10 radiologic findings (listhesis, annular fissure, disc bulge, disc extrusion, disc protrusion, endplate edema or Type 1 Modic change, lateral recess stenosis, Schmorl’s node, osteophyte, and any stenosis). In the seven potentially clinically important radiologic findings, the F1 score ranged from 0.9882 to 1.0. In this study, a rule-based NLP system identifying 23 findings related to LBP from X-ray, CT, and MRI reports was developed, and it presented good performance in regards to the four scoring parameters.

Abstract 13653: Leveraging Natural Language Processing and Machine Learning to Predict Worsening Heart Failure Events

Background: Prior risk models in patients with heart failure (HF) have focused on hospitalizations for worsening HF (WHF) and have not evaluated for differences in predictors by left ventricular ejection fraction (LVEF). We used natural language processing (NLP) and machine learning methods with access to longitudinal electronic health record (EHR) data to develop risk prediction models for WHF events across practice settings and by LVEF category. Methods: We identified all adults with HF and known LVEF on January 1 st of each year from 2011-2019 in an integrated health care system. WHF events within 1 year were defined as any hospitalization, emergency department, or outpatient encounter with ≥1 symptom, ≥2 objective findings including ≥1 sign, and ≥1 change in HF-related therapy. Signs and symptoms were ascertained using rule-based NLP. We conducted boosted decision tree-based ensemble models for any WHF event within each LVEF category: HF with reduced EF (HFrEF; LVEF ≤40%), HF with mildly reduced EF (HFmrEF; LVEF 41-49%), and HF with preserved EF (HFpEF; LVEF ≥50%). We evaluated model discrimination using area under the curve (AUC) and model calibration using Brier scores. Results: Among 359,298 patients from 2011-2019, 65,838 (18%) had HFrEF, 52,491 (15%) had HFmrEF, and 240,969 (67%) had HFpEF. Mean age was 75±12, 47% were women, and 37% were minorities including 10% Black, 11% Asian/Pacific Islander, and 12% of Hispanic ethnicity. WHF events occurred in 22% of patients with HFrEF, 17% with HFmrEF, and 16% with HFpEF. The models displayed an AUC of 0.75 and Brier score of 0.15 for HFrEF and an AUC of 0.77 and Brier scores of 0.12 for both HFmrEF and HFpEF. Clinical predictors were similar across LVEF categories ( Table ). Conclusions: Longitudinal EHR data can be leveraged using NLP and machine learning for accurate risk estimation that reliably identifies clinical predictors across a range of LVEF. These findings may provide novel insight into the natural history of HF.

Development and evaluation of an interoperable natural language processing system for identifying pneumonia across clinical settings of care and institutions.

To evaluate the feasibility, accuracy, and interoperability of a natural language processing (NLP) system that extracts diagnostic assertions of pneumonia in different clinical notes and institutions. A rule-based NLP system was designed to identify assertions of pneumonia in 3 types of clinical notes from electronic health records (EHRs): emergency department notes, radiology reports, and discharge summaries. The lexicon and classification logic were tailored for each note type. The system was first developed and evaluated using annotated notes from the Department of Veterans Affairs (VA). Interoperability was assessed using data from the University of Utah (UU). The NLP system was comprised of 782 rules and achieved moderate-to-high performance in all 3 note types in VA (precision/recall/f1: emergency = 88.1/86.0/87.1; radiology = 71.4/96.2/82.0; discharge = 88.3/93.0/90.1). When applied to UU data, performance was maintained in emergency and radiology but decreased in discharge summaries (emergency = 84.7/94.3/89.3; radiology = 79.7/100.0/87.9; discharge = 65.5/92.7/76.8). Customization with 34 additional rules increased performance for all note types (emergency = 89.3/94.3/91.7; radiology = 87.0/100.0/93.1; discharge = 75.0/95.1/83.4). NLP can be used to accurately identify the diagnosis of pneumonia across different clinical settings and institutions. A limited amount of customization to account for differences in lexicon, clinical definition of pneumonia, and EHR structure can achieve high accuracy without substantial modification.

Identifying Caregiver Availability Using Medical Notes With Rule-Based Natural Language Processing: Retrospective Cohort Study.

BackgroundIdentifying caregiver availability, particularly for patients with dementia or those with a disability, is critical to informing the appropriate care planning by the health systems, hospitals, and providers. This information is not readily available, and there is a paucity of pragmatic approaches to automatically identifying caregiver availability and type.ObjectiveOur main objective was to use medical notes to assess caregiver availability and type for hospitalized patients with dementia. Our second objective was to identify whether the patient lived at home or resided at an institution.MethodsIn this retrospective cohort study, we used 2016-2019 telephone-encounter medical notes from a single institution to develop a rule-based natural language processing (NLP) algorithm to identify the patient’s caregiver availability and place of residence. Using note-level data, we compared the results of the NLP algorithm with human-conducted chart abstraction for both training (749/976, 77%) and test sets (227/976, 23%) for a total of 223 adults aged 65 years and older diagnosed with dementia. Our outcomes included determining whether the patients (1) reside at home or in an institution, (2) have a formal caregiver, and (3) have an informal caregiver.ResultsTest set results indicated that our NLP algorithm had high level of accuracy and reliability for identifying whether patients had an informal caregiver (F1=0.94, accuracy=0.95, sensitivity=0.97, and specificity=0.93), but was relatively less able to identify whether the patient lived at an institution (F1=0.64, accuracy=0.90, sensitivity=0.51, and specificity=0.98). The most common explanations for NLP misclassifications across all categories were (1) incomplete or misspelled facility names; (2) past, uncertain, or undecided status; (3) uncommon abbreviations; and (4) irregular use of templates.ConclusionsThis innovative work was the first to use medical notes to pragmatically determine caregiver availability. Our NLP algorithm identified whether hospitalized patients with dementia have a formal or informal caregiver and, to a lesser extent, whether they lived at home or in an institutional setting. There is merit in using NLP to identify caregivers. This study serves as a proof of concept. Future work can use other approaches and further identify caregivers and the extent of their availability.

Development and validation of an automated basal cell carcinoma histopathology information extraction system using natural language processing.

Routinely collected healthcare data are a powerful research resource, but often lack detailed disease-specific information that is collected in clinical free text such as histopathology reports. We aim to use natural Language Processing (NLP) techniques to extract detailed clinical and pathological information from histopathology reports to enrich routinely collected data. We used the general architecture for text engineering (GATE) framework to build an NLP information extraction system using rule-based techniques. During validation, we deployed our rule-based NLP pipeline on 200 previously unseen, de-identified and pseudonymised basal cell carcinoma (BCC) histopathological reports from Swansea Bay University Health Board, Wales, UK. The results of our algorithm were compared with gold standard human annotation by two independent and blinded expert clinicians involved in skin cancer care. We identified 11,224 items of information with a mean precision, recall, and F1 score of 86.0% (95% CI: 75.1-96.9), 84.2% (95% CI: 72.8-96.1), and 84.5% (95% CI: 73.0-95.1), respectively. The difference between clinician annotator F1 scores was 7.9% in comparison with 15.5% between the NLP pipeline and the gold standard corpus. Cohen's Kappa score on annotated tokens was 0.85. Using an NLP rule-based approach for named entity recognition in BCC, we have been able to develop and validate a pipeline with a potential application in improving the quality of cancer registry data, supporting service planning, and enhancing the quality of routinely collected data for research.

Rule-based natural language processing for automation of stroke data extraction: a validation study.

Data extraction from radiology free-text reports is time consuming when performed manually. Recently, more automated extraction methods using natural language processing (NLP) are proposed. A previously developed rule-based NLP algorithm showed promise in its ability to extract stroke-related data from radiology reports. We aimed to externally validate the accuracy of CHARTextract, a rule-based NLP algorithm, to extract stroke-related data from free-text radiology reports. Free-text reports of CT angiography (CTA) and perfusion (CTP) studies of consecutive patients with acute ischemic stroke admitted to a regional stroke center for endovascular thrombectomy were analyzed from January 2015 to 2021. Stroke-related variables were manually extracted as reference standard from clinical reports, including proximal and distal anterior circulation occlusion, posterior circulation occlusion, presence of ischemia or hemorrhage, Alberta stroke program early CT score (ASPECTS), and collateral status. These variables were simultaneously extracted using a rule-based NLP algorithm. The NLP algorithm's accuracy, specificity, sensitivity, positive predictive value (PPV), and negative predictive value (NPV) were assessed. The NLP algorithm's accuracy was > 90% for identifying distal anterior occlusion, posterior circulation occlusion, hemorrhage, and ASPECTS. Accuracy was 85%, 74%, and 79% for proximal anterior circulation occlusion, presence of ischemia, and collateral status respectively. The algorithm confirmed the absence of variables from radiology reports with an 87-100% accuracy. Rule-based NLP has a moderate to good performance for stroke-related data extraction from free-text imaging reports. The algorithm's accuracy was affected by inconsistent report styles and lexicon among reporting radiologists.

Integrating corpus-based and NLP approach to extract terminology and domain-oriented information: an example of US military corpus

Within the modern information, communication and technology (ICT), seeking high efficient and accurate corpus-based approaches to process natural language data (NLD) is critical. Traditional corpus-based approaches for processing corpus (i.e. the collected NLD) mainly focused on quantifying and ranking words for assisting human in extracting keywords. However, traditional corpus-based approaches cannot identify the meanings behind the words to properly extract terminologies nor their information. To address this issue, the main objective of this paper is to propose an integrated linguistic analysis approach that combines two corpus-based approaches and a rule-based natural language processing (NLP) approach to extract and identify terminologies and create the text database for extracting deeper domain-oriented information by using the terminologies as channels to retrieve core information from the target corpus. Military domain is an uncommon research field and often classified as confidential data, which caused little researches to focus on. Nevertheless, military information is vital to national security and should not be ignored. Hence, to verify the proposed approach in extracting terminologies and information of the terminologies, the researchers adopt the US Army field manual (FM) 8-10-6 as the target corpus and empirical case. Compared with AntConc 3.5.8 and Tongpoon-Patanasorn’s hybrid approach, the results indicate that from the perspectives of terminology identification, texts database creation, domain knowledge extraction, only the proposed approach can handle all these issues.

A method to advance adolescent sexual health research: Automated algorithm finds sexual history documentation.

Background:We aimed to develop and validate a rule-based Natural Language Processing (NLP) algorithm to detect sexual history documentation and its five key components [partners, practices, past history of sexually transmitted infections (STIs), protection from STIs, and prevention of pregnancy] among adolescent encounters in the pediatric emergency and inpatient settings.MethodsWe iteratively designed a NLP algorithm using pediatric emergency department (ED) provider notes from adolescent ED visits with specific abdominal or genitourinary (GU) chief complaints. The algorithm is composed of regular expressions identifying commonly used phrases in sexual history documentation. We validated this algorithm with inpatient admission notes for adolescents. We calculated the sensitivity, specificity, negative predictive value, positive predictive value, and F1 score of the tool in each environment using manual chart review as the gold standard.ResultsIn the ED test cohort with abdominal or GU complaints, 97/179 (54%) provider notes had a sexual history documented, and the NLP algorithm correctly classified each note. In the inpatient validation cohort, 97/321 (30%) admission notes included a sexual history, and the NLP algorithm had 100% sensitivity and 98.2% specificity. The algorithm demonstrated >97% sensitivity and specificity in both settings for detection of elements of a high quality sexual history including protection used and contraception. Type of sexual practice and STI testing offered were also detected with >97% sensitivity and specificity in the ED test cohort with slightly lower performance in the inpatient validation cohort.ConclusionThis NLP algorithm automatically detects the presence of sexual history documentation and its key components in ED and inpatient settings.