Assisting radiologists with reporting urgent findings to referring physicians: A machine learning approach to identify cases for prompt communication

Xing, Meng; Ganoe, Craig H.; Sieberg, Ryan T.; Cheung, Yvonne; Hassanpour, Saeed

doi:10.1016/j.jbi.2019.103169

Cited by 18 publications

(11 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Information and communication technology may offer solutions to better organise and facilitate the reporting process [3,11,13]. Automatic detection of actionable findings using natural language processing may further support the radiologist in consistently detecting and reporting actionable findings [14][15][16][17]. Structured reporting also has potential value in reporting actionable findings [18].…”

Section: Discussionmentioning

confidence: 99%

Assessment of actionable findings in radiology reports

Visser

Vries

Kors

2020

European Journal of Radiology

View full text Add to dashboard Cite

Reporting Work Group defined three categories of imaging findings that require additional, nonroutine communication with the referring physician because of their urgency or unexpectedness. The objective of this study was to determine the prevalence of actionable findings in radiology reports, and to assess how well radiologists agree on the categorisation of actionable findings. Method: From 124,909 consecutive radiology reports stored in the electronic health record system of a large university hospital, 1000 reports were randomly selected. Two radiologists independently annotated all actionable findings according to the three categories of urgency defined by the ACR Work Group. Annotation differences were resolved in a consensus meeting and a final category was established for each report. Interannotator agreement was measured by accuracy and the kappa coefficient. Results: The prevalence of the three categories of actionable findings together was 32.5 %. Of all reports, 10.9 % were from patients seen in the emergency department. Prevalence of actionable findings for these patients (45.9 %) was considerably higher than for patients in routine clinical care (30.9 %). Interannotator agreement scores on the categorisation of actionable findings were 0.812 for accuracy and 0.616 for kappa coefficient. Conclusions: The prevalence of actionable findings in radiology reports is high. The interannotator agreement scores are moderate, indicating that categorisation of actionable findings is a difficult task. To avoid unneeded increase in the workload of radiologists, in particular in routine practice, clinical context may need to be considered in deciding whether a finding is actionable.

show abstract

Section: Discussionmentioning

confidence: 99%

Assessment of actionable findings in radiology reports

Visser

Vries

Kors

2020

European Journal of Radiology

View full text Add to dashboard Cite

show abstract

“…We are not the first to automatically detect actionable findings in radiology reports. Previous studies have reported different approaches, often with excellent results [2][3][4][5][6][7][8][9][10][11]. However, almost all of these studies focused on the detection of a limited set of findings, often related to a specific modality or organ system.…”

Section: Discussionmentioning

confidence: 99%

“…Monitoring the detection and communication of actionable findings in reports delivered in routine clinical care is important for quality control and auditing purposes, but requires laborious manual review that may not be feasible at scale. A considerable number of studies have used natural language processing to automate the detection of actionable findings in radiology reports [2][3][4][5][6][7][8][9][10][11]. However, almost all of these studies focused on the detection of a limited set of findings (often related to a specific modality or organ system) or even a single finding.…”

Section: Introductionmentioning

confidence: 99%

Automatic detection of actionable findings and communication mentions in radiology reports using natural language processing

2022

View full text Add to dashboard Cite

Objectives. To develop and validate classifiers for automatic detection of actionable findings and documentation of nonroutine communication in routinely delivered radiology reports. Methods.Two radiologists annotated all actionable findings and communication mentions in a training set of 1,306 radiology reports and a test set of 1,000 reports randomly selected from the electronic health record system of a large tertiary hospital. Various feature sets were constructed based on the impression section of the reports using different preprocessing steps (stemming, removal of stopwords, negations, and previously known or stable findings) and n-grams. Random forest classifiers were trained to detect actionable findings, and a decision-rule classifier was trained to find communication mentions. Classifier performance was evaluated by the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity.Results. On the training set, the actionable finding classifier with the highest cross-validated performance was obtained for a feature set of unigrams, after stemming and removal of negated, known and stable findings. On the test set, this classifier achieved an AUC of 0.876 (95% CI 0.854-0.898). The classifier for communication detection was trained after negation removal, using unigrams as features. The resultant decision rule had a sensitivity of 0.841 (95% CI 0.706-0.921) and specificity of 0.990 (95% CI 0.981-0.994) on the test set.Conclusions. Automatic detection of actionable findings and subsequent communication in routinely delivered radiology reports is possible. This can serve quality control purposes and may alert radiologists to the presence of actionable findings during reporting.

show abstract

“…This CDSS can assist the physicians in decision-making, by integrating clinical protocols and information regarding a specific patient. In [121], a semi-supervised NLP methodology was adopted to analyze the free-text narratives in the report with the aim of identifying patients with urgent radiological findings that require a rapid communication to their referring physicians. Similarly, Becker et al in [83] exploited an NLP analysis for patient-specific guidelines.…”

Section: Text Mining For Optimized Decision-makingmentioning

confidence: 99%

Recent advances of HCI in decision-making tasks for optimized clinical workflows and precision medicine

Rundo

Pirrone

Vitabile

et al. 2020

Journal of Biomedical Informatics

View full text Add to dashboard Cite

The ever-increasing amount of biomedical data is enabling new large-scale studies, even though ad hoc computational solutions are required. The most recent Machine Learning (ML) and Artificial Intelligence (AI) techniques have been achieving outstanding performance and an important impact in clinical research, aiming at precision medicine as well as improving healthcare workflows. However, the inherent heterogeneity and uncertainty in the healthcare information sources pose new compelling challenges for clinicians in their decision-making tasks. Only the proper combination of AI and human intelligence capabilities, by explicitly taking into account effective and safe interaction paradigms, can permit the delivery of care that outperforms what either can do separately. Therefore, Human-Computer Interaction (HCI) plays a crucial role in the design of software oriented to decision-making in medicine. In this work, we systematically review and discuss several research fields strictly linked to HCI and clinical decision-making, by subdividing the articles into six themes,

show abstract

Assisting radiologists with reporting urgent findings to referring physicians: A machine learning approach to identify cases for prompt communication

Cited by 18 publications

References 27 publications

Assessment of actionable findings in radiology reports

Assessment of actionable findings in radiology reports

Automatic detection of actionable findings and communication mentions in radiology reports using natural language processing

Recent advances of HCI in decision-making tasks for optimized clinical workflows and precision medicine

Contact Info

Product

Resources

About