Completeness of ultrasound reporting impacts time to biopsy for benign and malignant thyroid nodules

Inman, Alexandra; Liu, Kaidi; Ong, Kaye; Tiwari, Pari; Vos, Patrick M.; White, Adam; Wiseman, Sam M.

doi:10.1016/j.amjsurg.2017.03.030

Cited by 11 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, commenting on echogenicity in addition to other suspicious malignant features should be complementary and not mutually exclusive. Furthermore, since different studies favored different criteria as better predictors of malignancy, the more features reported the better is the distinction between benign and malignant lesions [1-3, 13, 14, 17, 19, 20]. In our study, the low rate of reporting on each of the suspicious malignant features and even lower cumulative reporting on multiple criteria is quite worrisome.…”

Section: Discussioncontrasting

confidence: 39%

“…Few studies have documented the suboptimal quality at some centers, in addition to a widespread dissatisfaction by endocrinologists and surgeons with these reports [15-20]. The negative impact of incomplete reporting on timely cytological diagnosis and surgical intervention has been documented [19, 20]. Nevertheless, Griffin et al [21] proved recently, in a three-phase study, that simple implementation of the ACR TI-RADS structured reporting template improved the quality significantly.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Thyroid Ultrasound Reports: Deficiencies and Recommendations

2019

View full text Add to dashboard Cite

Objective: Thyroid ultrasound plays a major role in the clinical management of patients with thyroid nodules. Comprehensive reporting enables malignant risk stratification and biopsy decisions. In this study, we aimed at a systematic evaluation of the content and completeness of thyroid radiology reports. Methods: A retrospective study was undertaken. A total of 200 thyroid ultrasound reports of examinations performed over a 1-year period were reviewed. After excluding 18 reports, the remaining 182 were evaluated for the inclusion of the following nodule characteristics: size, 3-axis dimensions, location, presence or absence of five signs suspicious of malignancy, namely microcalcification, hypoechogenecity, irregular margin, height-to-width ratio, and intranodular vascularity. Results: While all reported nodules could be stratified easily as being more or less than 1 cm in size, only 23.6% of these nodules were reported in 3 dimensions, and 33.5% of the nodules were specifically localized. For any described nodule, the frequency of reporting on echogenicity was 50%, on vascularity 19.2%, on margin 10.4%, on calcifications 9.3%, and no report contained a description of the height-to-width ratio. The cumulative frequency of reporting on one characteristic per nodule was 84%, of two characteristics 27%, three characteristics 4.4%, and no report included ≥4 characteristics per nodule. Conclusion: Despite easily accessible templates, reporting of thyroid nodule sonogram continues to be incomplete and inconsistent. This in turn constitutes a waste of a significant tool that could otherwise help in making timely informed medical decisions and in providing a significant platform for patients’ future follow-up.

show abstract

Section: Discussioncontrasting

confidence: 39%

Section: Introductionmentioning

confidence: 99%

Thyroid Ultrasound Reports: Deficiencies and Recommendations

2019

View full text Add to dashboard Cite

show abstract

“…These findings, which are in contrast to the work of Inman et al, may be due to the tendency of radiologists at the academic medical center in our study to include recommendations for clinical intervention. 18,31 Although the academic medical center had a slightly lower rate of completeness for all parameters, most individual parameters did not vary between the academic medical center and the regional health care system. Structured or synoptic reporting of thyroid US may improve both completeness of thyroid US reporting as well as difficulty with interpretation for human annotators.…”

Section: Discussionmentioning

confidence: 96%

Ambiguous and Incomplete: Natural Language Processing Reveals Problematic Reporting Styles in Thyroid Ultrasound Reports

et al. 2022

View full text Add to dashboard Cite

Objective Natural language processing (NLP) systems convert unstructured text into analyzable data. Here, we describe the performance measures of NLP to capture granular details on nodules from thyroid ultrasound (US) reports and reveal critical issues with reporting language. Methods We iteratively developed NLP tools using clinical Text Analysis and Knowledge Extraction System (cTAKES) and thyroid US reports from 2007 to 2013. We incorporated nine nodule features for NLP extraction. Next, we evaluated the precision, recall, and accuracy of our NLP tools using a separate set of US reports from an academic medical center (A) and a regional health care system (B) during the same period. Two physicians manually annotated each test-set report. A third physician then adjudicated discrepancies. The adjudicated “gold standard” was then used to evaluate NLP performance on the test-set. Results A total of 243 thyroid US reports contained 6,405 data elements. Inter-annotator agreement for all elements was 91.3%. Compared with the gold standard, overall recall of the NLP tool was 90%. NLP recall for thyroid lobe or isthmus characteristics was: laterality 96% and size 95%. NLP accuracy for nodule characteristics was: laterality 92%, size 92%, calcifications 76%, vascularity 65%, echogenicity 62%, contents 76%, and borders 40%. NLP recall for presence or absence of lymphadenopathy was 61%. Reporting style accounted for 18% errors. For example, the word “heterogeneous” interchangeably referred to nodule contents or echogenicity. While nodule dimensions and laterality were often described, US reports only described contents, echogenicity, vascularity, calcifications, borders, and lymphadenopathy, 46, 41, 17, 15, 9, and 41% of the time, respectively. Most nodule characteristics were equally likely to be described at hospital A compared with hospital B. Conclusions NLP can automate extraction of critical information from thyroid US reports. However, ambiguous and incomplete reporting language hinders performance of NLP systems regardless of institutional setting. Standardized or synoptic thyroid US reports could improve NLP performance.

show abstract

“…15 Underreporting of nodule characteristics as recommended by guidelines, leads to delayed diagnosis by cytology and delayed treatment. 16 Aiming to improve the standard of thyroid ultrasound reporting, a quality improvement study carried out in the United States in 2018 compared the standard of reports prior to and post-introduction of an ACR TI-RADS reporting guideline. It was found that the standard of ultrasound reports and definitive management recommendations improved significantly.…”

Section: Introductionmentioning

confidence: 99%