EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Uncertainty Estimation for Dual View X-ray Mammographic Image Registration Using Deep Ensembles

William C Walton, Seung-Jun Kim

Journal of Imaging Informatics in Medicine · 2024

DOI: 10.1007/s10278-024-01244-1 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging Informatics in Medicine, 2024. Verfasst von William C Walton und 1 weiteren Autor:innen. Beteiligte Einrichtung: University of Maryland, Baltimore County, CSEE Department, Baltimore, MD, 21250, USA. Hauptthemen laut MEDLINE: Mammography, Breast Neoplasms, Neural Networks, Computer, Radiographic Image Interpretation, Computer-Assisted. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Techniques are developed for generating uncertainty estimates for convolutional neural network (CNN)-based methods for registering the locations of lesions between the craniocaudal (CC) and mediolateral oblique (MLO) mammographic X-ray image views. Multi-view lesion correspondence is an important task that clinicians perform for characterizing lesions during routine mammographic exams. Automated registration tools can aid in this task, yet if the tools also provide confidence estimates, they can be of greater value to clinicians, especially in cases involving dense tissue where lesions may be difficult to see. A set of deep ensemble-based techniques, which leverage a negative log-likelihood (NLL)-based cost function, are implemented for estimating uncertainties. The ensemble architectures involve significant modifications to an existing CNN dual-view lesion registration algorithm. Three architectural designs are evaluated, and different ensemble sizes are compared using various performance metrics. The techniques are tested on synthetic X-ray data, real 2D X-ray data, and slices from real 3D X-ray data. The ensembles generate covariance-based uncertainty ellipses that are correlated with registration accuracy, such that the ellipse sizes can give a clinician an indication of confidence in the mapping between the CC and MLO views. The results also show that the ellipse sizes can aid in improving computer-aided detection (CAD) results by matching CC/MLO lesion detects and reducing false alarms from both views, adding to clinical utility. The uncertainty estimation techniques show promise as a means for aiding clinicians in confidently establishing multi-view lesion correspondence, thereby improving diagnostic capability.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MammographyBreast NeoplasmsNeural Networks, ComputerRadiographic Image Interpretation, Computer-AssistedDeep LearningHumansUncertaintyFemaleAlgorithms

Stichwörter der Autor:innen

Breast cancerImage registrationLesion correspondenceMammographyNeural networkUncertainty

Autor:innen und Institutionen

  • University of Maryland, Baltimore County, CSEE Department, Baltimore, MD, 21250, USA.
  • The Johns Hopkins University Applied Physics Laboratory, Laurel, MD, 20723, USA.
  • University of Maryland, Baltimore County, CSEE Department, Baltimore, MD, 21250, USA. sjkim@umbc.edu.

Zitieren

APA

Walton, W. C., & Kim, S. J. (2024). Uncertainty Estimation for Dual View X-ray Mammographic Image Registration Using Deep Ensembles. Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-024-01244-1

Vancouver

Walton WC, Kim SJ. Uncertainty Estimation for Dual View X-ray Mammographic Image Registration Using Deep Ensembles. Journal of Imaging Informatics in Medicine. 2024. doi:10.1007/s10278-024-01244-1

BibTeX
@article{Walton2024,
  author = {Walton, William C and Kim, Seung-Jun},
  title = {Uncertainty Estimation for Dual View X-ray Mammographic Image Registration Using Deep Ensembles},
  journal = {Journal of Imaging Informatics in Medicine},
  year = {2024},
  doi = {10.1007/s10278-024-01244-1}
}
Kostenlos starten →