Dataset Information

Natural Language Processing for Identification of Incidental Pulmonary Nodules in Radiology Reports.

ABSTRACT: PURPOSE:To develop natural language processing (NLP) to identify incidental lung nodules (ILNs) in radiology reports for assessment of management recommendations. METHODS AND MATERIALS:We searched the electronic health records for patients who underwent chest CT during 2014 and 2017, before and after implementation of a department-wide dictation macro of the Fleischner Society recommendations. We randomly selected 950 unstructured chest CT reports and reviewed manually for ILNs. An NLP tool was trained and validated against the manually reviewed set, for the task of automated detection of ILNs with exclusion of previously known or definitively benign nodules. For ILNs found in the training and validation sets, we assessed whether reported management recommendations agreed with Fleischner Society guidelines. The guideline concordance of management recommendations was compared between 2014 and 2017. RESULTS:The NLP tool identified ILNs with sensitivity and specificity of 91.1% and 82.2%, respectively, in the validation set. Positive and negative predictive values were 59.7% and 97.0%. In reports of ILNs in the training and validation sets before versus after introduction of a Fleischner reporting macro, there was no difference in the proportion of reports with ILNs (108 of 500 [21.6%] versus 101 of 450 [22.4%]; P = .8), or in the proportion of reports with ILNs containing follow-up recommendations (75 of 108 [69.4%] versus 80 of 101 [79.2%]; P = .2]. Rates of recommendation guideline concordance were not significantly different before and after implementation of the standardized macro (52 of 75 [69.3%] versus 60 of 80 [75.0%]; P = .43). CONCLUSION:NLP reliably automates identification of ILNs in unstructured reports, pertinent to quality improvement efforts for ILN management.

SUBMITTER: Kang SK

PROVIDER: S-EPMC7686874 | biostudies-literature | 2019 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Natural Language Processing for Identification of Incidental Pulmonary Nodules in Radiology Reports.

Kang Stella K SK Garry Kira K Chung Ryan R Moore William H WH Iturrate Eduardo E Swartz Jordan L JL Kim Danny C DC Horwitz Leora I LI Blecker Saul S

Journal of the American College of Radiology : JACR 20190524 11

<h4>Purpose</h4>To develop natural language processing (NLP) to identify incidental lung nodules (ILNs) in radiology reports for assessment of management recommendations.<h4>Methods and materials</h4>We searched the electronic health records for patients who underwent chest CT during 2014 and 2017, before and after implementation of a department-wide dictation macro of the Fleischner Society recommendations. We randomly selected 950 unstructured chest CT reports and reviewed manually for ILNs. A ...[more]

PMID: 31132331

Similar Datasets

Project description:BackgroundNatural language processing (NLP) has a significant role in advancing healthcare and has been found to be key in extracting structured information from radiology reports. Understanding recent developments in NLP application to radiology is of significance but recent reviews on this are limited. This study systematically assesses and quantifies recent literature in NLP applied to radiology reports.MethodsWe conduct an automated literature search yielding 4836 results using automated filtering, metadata enriching steps and citation search combined with manual review. Our analysis is based on 21 variables including radiology characteristics, NLP methodology, performance, study, and clinical application characteristics.ResultsWe present a comprehensive analysis of the 164 publications retrieved with publications in 2019 almost triple those in 2015. Each publication is categorised into one of 6 clinical application categories. Deep learning use increases in the period but conventional machine learning approaches are still prevalent. Deep learning remains challenged when data is scarce and there is little evidence of adoption into clinical practice. Despite 17% of studies reporting greater than 0.85 F1 scores, it is hard to comparatively evaluate these approaches given that most of them use different datasets. Only 14 studies made their data and 15 their code available with 10 externally validating results.ConclusionsAutomated understanding of clinical narratives of the radiology reports has the potential to enhance the healthcare process and we show that research in this field continues to grow. Reproducibility and explainability of models are important if the domain is to move applications into clinical use. More could be done to share code enabling validation of methods on different institutional data and to reduce heterogeneity in reporting of study properties allowing inter-study comparisons. Our results have significance for researchers in the field providing a systematic synthesis of existing work to build on, identify gaps, opportunities for collaboration and avoid duplication.

Project description:ImportanceA rapid learning health care system for oncology will require scalable methods for extracting clinical end points from electronic health records (EHRs). Outside of clinical trials, end points such as cancer progression and response are not routinely encoded into structured data.ObjectiveTo determine whether deep natural language processing can extract relevant cancer outcomes from radiologic reports, a ubiquitous but unstructured EHR data source.Design, setting, and participantsA retrospective cohort study evaluated 1112 patients who underwent tumor genotyping for a diagnosis of lung cancer and participated in the Dana-Farber Cancer Institute PROFILE study from June 26, 2013, to July 2, 2018.ExposuresPatients were divided into curation and reserve sets. Human abstractors applied a structured framework to radiologic reports for the curation set to ascertain the presence of cancer and changes in cancer status over time (ie, worsening/progressing vs improving/responding). Deep learning models were then trained to capture these outcomes from report text and subsequently evaluated in a 10% held-out test subset of curation patients. Cox proportional hazards regression models compared human and machine curations of disease-free survival, progression-free survival, and time to improvement/response in the curation set, and measured associations between report classification and overall survival in the curation and reserve sets.Main outcomes and measuresThe primary outcome was area under the receiver operating characteristic curve (AUC) for deep learning models; secondary outcomes were time to improvement/response, disease-free survival, progression-free survival, and overall survival.ResultsA total of 2406 patients were included (mean [SD] age, 66.5 [10.8] years; 1428 female [59.7%]; 2170 [90.2%] white). Radiologic reports (n = 14 230) were manually reviewed for 1112 patients in the curation set. In the test subset (n = 109), deep learning models identified the presence of cancer, improvement/response, and worsening/progression with accurate discrimination (AUC >0.90). Machine and human curation yielded similar measurements of disease-free survival (hazard ratio [HR] for machine vs human curation, 1.18; 95% CI, 0.71-1.95); progression-free survival (HR, 1.11; 95% CI, 0.71-1.71); and time to improvement/response (HR, 1.03; 95% CI, 0.65-1.64). Among 15 000 additional reports for 1294 reserve set patients, algorithm-detected cancer worsening/progression was associated with decreased overall survival (HR for mortality, 4.04; 95% CI, 2.78-5.85), and improvement/response was associated with increased overall survival (HR, 0.41; 95% CI, 0.22-0.77).Conclusions and relevanceDeep natural language processing appears to speed curation of relevant cancer outcomes and facilitate rapid learning from EHR data.

Project description:Study aimTo develop and apply a natural language processing algorithm for characterization of patients diagnosed with chronic pancreatitis in a diverse integrated U.S. healthcare system.MethodsRetrospective cohort study including patients initially diagnosed with chronic pancreatitis (CP) within a regional integrated healthcare system between January 1, 2006 and December 31, 2015. Imaging reports from these patients were extracted from the electronic medical record system and split into training, validation and implementation datasets. A natural language processing (NLP) algorithm was first developed through the training dataset to identify specific features (atrophy, calcification, pseudocyst, cyst and main duct dilatation) from free-text radiology reports. The validation dataset was applied to validate the performance by comparing against the manual chart review. The developed algorithm was then applied to the implementation dataset. We classified patients with calcification(s) or ≥2 radiographic features as advanced CP. We compared etiology, comorbid conditions, treatment parameters as well as survival between advanced CP and others diagnosed during the study period.Results6,346 patients were diagnosed with CP during the study period with 58,085 radiology studies performed. For individual features, NLP yielded sensitivity from 88.7% to 95.3%, specificity from 98.2% to 100.0%. A total of 3,672 patients met cohort inclusion criteria: 1,330 (36.2%) had evidence of advanced CP. Patients with advanced CP had increased frequency of smoking (57.8% vs. 43.0%), diabetes (47.6% vs. 35.9%) and underweight body mass index (6.6% vs. 3.6%), all p<0.001. Mortality from pancreatic cancer was higher in advanced CP (15.3/1,000 person-year vs. 2.8/1,000, p<0.001). Underweight BMI (HR 1.6, 95% CL 1.2, 2.1), smoking (HR 1.4, 95% CL 1.1, 1.7) and diabetes (HR 1.4, 95% CL 1.2, 1.6) were independent risk factors for mortality.ConclusionPatients with advanced CP experienced increased disease-related complications and pancreatic cancer-related mortality. Excess all-cause mortality was driven primarily by potentially modifiable risk factors including malnutrition, smoking and diabetes.

Project description:Background and purposeVariability in radiologists' reporting styles and recommendations for incidental thyroid nodules can lead to confusion among clinicians and may contribute to inconsistent patient care. Our aim was to describe reporting practices of radiologists for incidental thyroid nodules seen on CT and MR imaging and to determine factors that influence reporting styles.Materials and methodsThis is a retrospective study of patients with incidental thyroid nodules reported on CT and MR imaging between January and December 2011, identified by text search for "thyroid nodule" in all CT and MR imaging reports. The studies included CT and MR imaging scans of the neck, spine, and chest. Radiology reports were divided into those that mentioned the incidental thyroid nodules only in the "Findings" section versus those that reported the incidental thyroid nodules in the "Impression" section as well, because this latter reporting style gives more emphasis to the finding. Univariate and multivariate analyses were performed to identify radiologist, patient, and nodule characteristics that influenced reporting styles.ResultsThree hundred seventy-five patients met the criterion of having incidental thyroid nodules. One hundred thirty-eight (37%) patients had incidental thyroid nodules reported in the "Impression" section. On multivariate analysis, only radiologists' divisions and nodule size were associated with reporting in "Impression." Chest radiologists and neuroradiologists were more likely to report incidental thyroid nodules in the "Impression" section than their abdominal imaging colleagues, and larger incidental thyroid nodules were more likely to be reported in "Impression" (P ≤ .03). Seventy-three percent of patients with incidental thyroid nodules of ≥20 mm were reported in the "Impression" section, but higher variability in reporting was seen for incidental thyroid nodules measuring 10-14 mm and 15-19 mm, which were reported in "Impression" for 61% and 50% of patients, respectively.ConclusionsReporting practices for incidental thyroid nodules detected on CT and MR imaging are predominantly influenced by nodule size and the radiologist's subspecialty. Reporting was highly variable for nodules measuring 10-19 mm; this finding can be partially attributed to different reporting styles among radiology subspecialty divisions. The variability demonstrated in this study further underscores the need to develop CT and MR imaging practice guidelines with the goal of standardizing reporting of incidental thyroid nodules and thereby potentially improving the consistency and quality of patient care.

Dataset Information

Natural Language Processing for Identification of Incidental Pulmonary Nodules in Radiology Reports.

Publications

Natural Language Processing for Identification of Incidental Pulmonary Nodules in Radiology Reports.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets