• Corpus ID: 18002058

Identifying Significance of Discrepancies in Radiology Reports

@inproceedings{Cohan2016IdentifyingSO,
  title={Identifying Significance of Discrepancies in Radiology Reports},
  author={Arman Cohan and Luca Soldaini and Nazli Goharian and Allan Fong and Ross W. Filice and Raj M. Ratwani},
  year={2016}
}
At many teaching hospitals, it is common practice for on-call radiology residents to interpret radiology examinations; such reports are later reviewed and revised by an attending physician before being used for any decision making. In case there are substantial problems in the resident’s initial report, the resident is called and the problems are reviewed to prevent similar future reporting errors. However, due to the large volume of reports produced, attending physicians rarely discuss the… 

Figures and Tables from this paper

Ranking Significant Discrepancies in Clinical Reports
TLDR
A novel ranking approach, consisting of textual and ontological overlaps between the preliminary and final versions of reports, is proposed, which allows medical practitioners to easily identify and learn from the reports in which their interpretation most substantially differed from that of the attending physician.
Domain Category Information as a Guide for Sentence Ranking to Support Medical Text Summarization
TLDR
This thesis contributes to existing research by proposing sentence ranking strategies to incorporate domain-specific categorial information to support medical text summarization.
The National Academies Board on Human System Integration (BOHSI) Panel: Promise, Progress and Challenges of Leveraging AI Technology in Healthcare
TLDR
The Panel brings together prominent researchers from both the Human Factors and Computer Science and AI communities to discuss promising applications, challenges and short-falls and ways forward for achieving the promise of enhanced delivery of health care through application of AI technologies.
Text Summarization and Categorization for Scientific and Health-Related Data
TLDR
This research goal in this dissertation is to develop Natural Language Processing and Information Retrieval methods for better processing and understanding health-related textual information to promote health care and wellbeing of individuals.

References

SHOWING 1-10 OF 32 REFERENCES
Identifying benchmarks for discrepancy rates in preliminary interpretations provided by radiology trainees at an academic institution.
Radiology Report Comparator: A Novel Method to Augment Resident Education
TLDR
This work created an electronic method for retrieving and displaying changes made to resident written preliminary reports by attending radiologists during the process of radiology report finalization and found residents compared reports more frequently after deployment of this software solution.
The DePICTORS Study: discrepancies in preliminary interpretation of CT scans between on-call residents and staff
TLDR
A low discrepancy rate between residents and staff radiologists is demonstrated and areas where after-hours service may be further improved are identified.
The discrepancy rate between preliminary and official reports of emergency radiology studies: a performance indicator and quality improvement method
TLDR
A system for follow-up of discrepancy in interpreting emergency radiology studies, and a periodical analysis might help to reduce the number of discrepant reports by targeted intervention is demonstrated.
Comparison-Bot: an Automated Preliminary-Final Report Comparison System
TLDR
This work developed a system that automatically compiles and emails a weekly summary of report differences for all residents and fellows and provides an easy way for trainees to review changes to preliminary reports with immediate access to the associated images, thereby improving their educational experience.
Automatic selection of radiological protocols using machine learning
TLDR
An ontology built by the National Library of Medicine is exploited to provide domain expertise, as well as an associated parser that maps free-form text into this ontology, using a graph-based semi-supervised learning technique.
Research and applications: Supervised machine learning and active learning in classification of radiology reports
TLDR
An automated system for classifying the results of imaging examinations into reportable and non-reportable cancer cases that can dramatically reduce human effort in identifying relevant reports from the large imaging pool for further investigation of cancer.
Research Paper: Creating a Text Classifier to Detect Radiology Reports Describing Mediastinal Findings Associated with Inhalational Anthrax and Other Disorders
TLDR
The IPS system is a useful tool for helping domain experts create a statistical key word classifier for textual reports that is a potentially useful component in surveillance of radiographic findings suspicious for anthrax.
Mixing Domain Rules with Machine Learning for Radiology Text Classification
TLDR
This work proposes a hybrid classication mechanism for radiology text combining rules and NLP/ML that in the authors' trials surpasses existing rule-based and N LP/ML techniques for the task of classifying incidental ndings.
...
...