tbiExtractor: A framework for extracting traumatic brain injury common data elements from radiology reports

Mahan, Margaret Y.; Rafter, Daniel; Casey, Hannah; Engelking, Marta; Abdallah, Tessneem; Truwit, Charles L.; Oswood, Mark; Samadani, Uzma

doi:10.1371/journal.pone.0214775

Cited by 7 publications

(3 citation statements)

References 36 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further development of these algorithms could potentially reduce implicit biases in the management of brain injury and improve outcomes for all patients, although great care must be taken to make sure that the algorithms are themselves not biased. Examples of assessors to include in these prognosticating algorithms include measures of brainstem function such as pupillometry, eye tracking or other quantitative cranial nerve function (30-33), serum markers (26), and image analysis (36)(37)(38)(39). These are measures that should potentially be able to be confirmed as "colorblind.…”

Section: Can Artificial Intelligence or Advanced Automation Correct T...mentioning

confidence: 99%

Consequences of inequity in the neurosurgical workforce: Lessons from traumatic brain injury

Venkatesh

Bravo²,

Schaaf³

et al. 2022

Front. Surg.

Self Cite

View full text Add to dashboard Cite

Women and minorities leave or fail to advance in the neurosurgical workforce more frequently than white men at all levels from residency to academia. The consequences of this inequity are most profound in fields such as traumatic brain injury (TBI), which lacks objective measures. We evaluated published articles on TBI clinical research and found that TBI primary investigators or corresponding authors were 86·5% White and 59·5% male. First authors from the resulting publications were 92.6% white. Most study participants were male (68%). 64·4% of NIH-funded TBI clinical trials did not report or recruit any black subjects and this number was even higher for other races and the Hispanic ethnicity. We propose several measures for mitigation of the consequences of the inequitable workforce in traumatic brain injury that could potentially contribute to more equitable outcomes. The most immediately feasible of these is validation and establishment of objective measures for triage and prognostication that are less susceptible to bias than current protocols. We call for incorporation of gender and race neutral metrics for TBI evaluation to standardize classification of injury. We offer insights into how socioeconomic factors contribute to increased death rates from women and minority groups. We propose the need to study how these disparities are caused by unfair health insurance reimbursement practices. Surgical and clinical research inequities have dire consequences, and until those inequities can be corrected, mitigation of those consequences requires system wide change.

show abstract

Section: Can Artificial Intelligence or Advanced Automation Correct T...mentioning

confidence: 99%

Consequences of inequity in the neurosurgical workforce: Lessons from traumatic brain injury

Venkatesh

Bravo²,

Schaaf³

et al. 2022

Front. Surg.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ten studies use NLP to create specific cohorts for research purposes and six reported the performance of their tools. Out of these papers, the majority (n=8) created cohorts for specific medical conditions including fatty liver disease [Goldshtein et al, 2020, Redman et al, 2017 hepatocellular cancer [Sada et al, 2016], ureteric stones [Li and Elliot, 2019], vertebral facture [Tan and Heagerty, 2019], traumatic brain injury [Yadav et al, 2016, Mahan et al, 2019, and leptomeningeal disease secondary to metastatic breast cancer [Brizzi et al, 2019]. Five papers identified cohorts focused on particular radiology findings including ground glass opacities (GGO) [Van Haren et al, 2019], cerebral microbleeds (CMB) [Noorbakhsh-Sabet et al, 2018], pulmonary nodules [Gould et al, 2015], [Huhdanpaa et al, 2018], changes in the spine correlated to back pain [Bates et al, 2016] and identifying radiological evidence of people having suffered a fall.…”

Section: Cohort and Epidemiologymentioning

confidence: 99%

A Systematic Review of Natural Language Processing Applied to Radiology Reports

Casey,

Davidson,

Poon

et al. 2021

Preprint

View full text Add to dashboard Cite

Background Natural 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.Methods We conduct an automated literature search yielding 4,799 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.Results We 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.Conclusions Automated 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

show abstract

“…Natural language processing (NLP) can support the automated reading of radiology reports ( 1 , 2 ). Research on clinical NLP has focused on improving the phenotyping of coded health data routinely collected during healthcare visits ( 3 ), enhancing cohort identification for research studies ( 4 , 5 ), and improving healthcare quality ( 6 , 7 ). However, a gap in translation from research to clinical application remains.…”

Section: Introductionmentioning

confidence: 99%

Understanding the performance and reliability of NLP tools: a comparison of four NLP tools predicting stroke phenotypes in radiology reports

Casey,

Davidson,

Grover

et al. 2023

Front. Digit. Health

View full text Add to dashboard Cite

BackgroundNatural language processing (NLP) has the potential to automate the reading of radiology reports, but there is a need to demonstrate that NLP methods are adaptable and reliable for use in real-world clinical applications.MethodsWe tested the F1 score, precision, and recall to compare NLP tools on a cohort from a study on delirium using images and radiology reports from NHS Fife and a population-based cohort (Generation Scotland) that spans multiple National Health Service health boards. We compared four off-the-shelf rule-based and neural NLP tools (namely, EdIE-R, ALARM+, ESPRESSO, and Sem-EHR) and reported on their performance for three cerebrovascular phenotypes, namely, ischaemic stroke, small vessel disease (SVD), and atrophy. Clinical experts from the EdIE-R team defined phenotypes using labelling techniques developed in the development of EdIE-R, in conjunction with an expert researcher who read underlying images.ResultsEdIE-R obtained the highest F1 score in both cohorts for ischaemic stroke, ≥93%, followed by ALARM+, ≥87%. The F1 score of ESPRESSO was ≥74%, whilst that of Sem-EHR is ≥66%, although ESPRESSO had the highest precision in both cohorts, 90% and 98%. For F1 scores for SVD, EdIE-R scored ≥98% and ALARM+ ≥90%. ESPRESSO scored lowest with ≥77% and Sem-EHR ≥81%. In NHS Fife, F1 scores for atrophy by EdIE-R and ALARM+ were 99%, dropping in Generation Scotland to 96% for EdIE-R and 91% for ALARM+. Sem-EHR performed lowest for atrophy at 89% in NHS Fife and 73% in Generation Scotland. When comparing NLP tool output with brain image reads using F1 scores, ALARM+ scored 80%, outperforming EdIE-R at 66% in ischaemic stroke. For SVD, EdIE-R performed best, scoring 84%, with Sem-EHR 82%. For atrophy, EdIE-R and both ALARM+ versions were comparable at 80%.ConclusionsThe four NLP tools show varying F1 (and precision/recall) scores across all three phenotypes, although more apparent for ischaemic stroke. If NLP tools are to be used in clinical settings, this cannot be performed “out of the box.” It is essential to understand the context of their development to assess whether they are suitable for the task at hand or whether further training, re-training, or modification is required to adapt tools to the target task.

show abstract

tbiExtractor: A framework for extracting traumatic brain injury common data elements from radiology reports

Cited by 7 publications

References 36 publications

Consequences of inequity in the neurosurgical workforce: Lessons from traumatic brain injury

Consequences of inequity in the neurosurgical workforce: Lessons from traumatic brain injury

A Systematic Review of Natural Language Processing Applied to Radiology Reports

Understanding the performance and reliability of NLP tools: a comparison of four NLP tools predicting stroke phenotypes in radiology reports

Contact Info

Product

Resources

About