Diagnostic Performance of SRU and ATA Thyroid Nodule Classification Algorithms as Tested With a 1 Million Virtual Thyroid Nodule Model.
Purpose: The Society of Radiologists in Ultrasound (SRU 2005) and American Thyroid Association (ATA 2009 and ATA 2015) have published algorithms regarding thyroid nodule management. Kwak et al. and other groups have described models that estimate thyroid nodules' malignancy risk. The aim of our stud...
| Publicado en: | Current Problems in Diagnostic Radiology Vol. 47; no. 1; pp. 10 - 14 |
|---|---|
| Autores principales: | , , , |
| Formato: | research Journal Article |
| Publicado: |
Elsevier B.V.
Jan2018
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=125943248&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 125943248 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 03630188 1YCI jtl: Current Problems in Diagnostic Radiology issn: 03630188 maglogo: N pubinfo: dt: Jan2018 vid: 47 iid: 1 pid: 467 pub: Elsevier B.V. place: New York, New York artinfo: ui: 125943248 125943248 NLM28554789 125943248 10.1067/j.cpradiol.2017.04.005 NLM28554789 125943248 ppf: 10 ppct: 4 formats: tig: atl: Diagnostic Performance of SRU and ATA Thyroid Nodule Classification Algorithms as Tested With a 1 Million Virtual Thyroid Nodule Model. aug: au: Boehnke, Mitchell Patel, Nayana McKinney, Kristin Clark, Toshimasa affil: Division of Abdominal Imaging, Department of Diagnostic Radiology, The University of Colorado Anschutz Medical Campus, Aurora, CO sug: subj: Classification Algorithms Thyroid Nodule Thyroid Neoplasms Thyroid Neoplasms Pathology Diagnosis, Computer Assisted Methods Thyroid Nodule Pathology Medical Organizations Sensitivity and Specificity Risk Assessment Diagnosis, Differential Scales Human ab: Purpose: The Society of Radiologists in Ultrasound (SRU 2005) and American Thyroid Association (ATA 2009 and ATA 2015) have published algorithms regarding thyroid nodule management. Kwak et al. and other groups have described models that estimate thyroid nodules' malignancy risk. The aim of our study is to use Kwak's model to evaluate the tradeoffs of both sensitivity and specificity of SRU 2005, ATA 2009 and ATA 2015 management algorithms.Materials and Methods: 1,000,000 thyroid nodules were modeled in MATLAB. Ultrasound characteristics were modeled after published data. Malignancy risk was estimated per Kwak's model and assigned as a binary variable. All nodules were then assessed using the published management algorithms. With the malignancy variable as condition positivity and algorithms' recommendation for FNA as test positivity, diagnostic performance was calculated.Results: Modeled nodule characteristics mimic those of Kwak et al. 12.8% nodules were assigned as malignant (malignancy risk range of 2.0-98%). FNA was recommended for 41% of nodules by SRU 2005, 66% by ATA 2009, and 82% by ATA 2015. Sensitivity and specificity is significantly different (< 0.0001): 49% and 60% for SRU; 81% and 36% for ATA 2009; and 95% and 20% for ATA 2015.Conclusion: SRU 2005, ATA 2009 and ATA 2015 algorithms are used routinely in clinical practice to determine whether thyroid nodule biopsy is indicated. We demonstrate significant differences in these algorithms' diagnostic performance, which result in a compromise between sensitivity and specificity. pubtype: Academic Journal doctype: research Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|