EUVIMEDEUVIMED entdecken
EvaluationsstudieMultizentrische Studie

Multi-institutional development and testing of attention-enhanced deep learning segmentation of thyroid nodules on ultrasound

Joseph L Cozzi, Hui Li, Jordan D Fuhrman, Li Lan, Jelani Williams, Brendan Finnerty, Thomas J Fahey, Abhinay Tumati, Joshua Genender, Xavier M Keutgen, Maryellen L Giger

International Journal of Computer Assisted Radiology and Surgery · 2025

DOI: 10.1007/s11548-024-03294-w ↗

Kurz gefasst

Evaluationsstudie in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, 2025. Verfasst von Joseph L Cozzi und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, University of Chicago, Chicago, IL, USA. jlcozzi@uchicago.edu. Hauptthemen laut MEDLINE: Deep Learning, Image Interpretation, Computer-Assisted, Thyroid Nodule. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 32 Literaturangaben. Gefördert von: NCATS NIH HHS, NIH HHS, National Center for Advancing Translational Sciences.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: Thyroid nodules are common, and ultrasound-based risk stratification using ACR's TIRADS classification is a key step in predicting nodule pathology. Determining thyroid nodule contours is necessary for the calculation of TIRADS scores and can also be used in the development of machine learning nodule diagnosis systems. This paper presents the development, validation, and multi-institutional independent testing of a machine learning system for the automatic segmentation of thyroid nodules on ultrasound. METHODS: The datasets, containing a total of 1595 thyroid ultrasound images from 520 patients with thyroid nodules, were retrospectively collected under IRB approval from University of Chicago Medicine (UCM) and Weill Cornell Medical Center (WCMC). Nodules were manually contoured by a team of UCM and WCMC physicians for ground truth. An AttU-Net, a U-Net architecture with additional attention weighting functions, was trained for the segmentations. The algorithm was validated through fivefold cross-validation by nodule and was tested on two independent test sets: one from UCM and one from WCMC. Dice similarity coefficient (DSC) and percent Hausdorff distance (%HD), Hausdorff distance reported as a percent of the nodule's effective diameter, served as the performance metrics. RESULTS: On multi-institutional independent testing, the AttU-Net yielded average DSCs (std. deviation) of 0.915 (0.04) and 0.922 (0.03) and %HDs (std. deviation) of 12.9% (4.6) and 13.4% (6.3) on the UCM and WCMC test sets, respectively. Similarity testing showed the algorithm's performance on the two institutional test sets was equivalent up to margins of Δ DSC ≤ 0.013 and Δ %HD ≤ 1.73%. CONCLUSIONS: This work presents a robust automatic thyroid nodule segmentation algorithm that could be implemented for risk stratification systems. Future work is merited to incorporate this segmentation method within an automatic thyroid classification system.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningImage Interpretation, Computer-AssistedThyroid NoduleFemaleHumansMaleAlgorithmsRetrospective StudiesUltrasonography

Stichwörter der Autor:innen

AttentionDeep learningNodulesSegmentationThyroid

Autor:innen und Institutionen

  • Department of Radiology, University of Chicago, Chicago, IL, USA. jlcozzi@uchicago.edu.
  • Department of Radiology, University of Chicago, Chicago, IL, USA.
  • Division of General Surgery and Surgical Oncology, Department of Surgery, University of Chicago Medicine, Endocrine Surgery Research Program, Chicago, IL, USA.
  • Endocrine Oncology Research Program, Division of Endocrine Surgery, Department of Surgery, New York Presbyterian Hospital-Weill Cornell Medicine, New York, USA.
  • David Geffen School of Medicine, University of California - Los Angeles, Los Angeles, CA, USA.

Förderung

  • NCATS NIH HHS
  • NIH HHS
  • National Center for Advancing Translational Sciences
  • National Institutes of Health

Zitieren

APA

Cozzi, J. L., Li, H., Fuhrman, J. D., Lan, L., Williams, J., Finnerty, B., Fahey, T. J., Tumati, A., Genender, J., Keutgen, X. M., & Giger, M. L. (2025). Multi-institutional development and testing of attention-enhanced deep learning segmentation of thyroid nodules on ultrasound. International Journal of Computer Assisted Radiology and Surgery. https://doi.org/10.1007/s11548-024-03294-w

Vancouver

Cozzi JL, Li H, Fuhrman JD, Lan L, Williams J, Finnerty B, et al. Multi-institutional development and testing of attention-enhanced deep learning segmentation of thyroid nodules on ultrasound. International Journal of Computer Assisted Radiology and Surgery. 2025. doi:10.1007/s11548-024-03294-w

BibTeX
@article{Cozzi2025,
  author = {Cozzi, Joseph L and Li, Hui and Fuhrman, Jordan D and Lan, Li and Williams, Jelani and Finnerty, Brendan and Fahey, Thomas J and Tumati, Abhinay and Genender, Joshua and Keutgen, Xavier M and Giger, Maryellen L},
  title = {Multi-institutional development and testing of attention-enhanced deep learning segmentation of thyroid nodules on ultrasound},
  journal = {International Journal of Computer Assisted Radiology and Surgery},
  year = {2025},
  doi = {10.1007/s11548-024-03294-w}
}
Kostenlos starten →