Incremental retraining, clinical implementation, and acceptance rate of deep learning auto-segmentation for male pelvis in a multiuser environment.

Journal: Medical physics
Published Date:

Abstract

BACKGROUND: Deep learning auto-segmentation (DLAS) models have been adopted in the clinic; however, they suffer from performance deterioration owing to the clinical practice variability. Some commercial DLAS software provide an incremental retraining function that enables users to train a custom model using their institutional data to account for clinical practice variability.

Authors

  • Jingwei Duan
    Department of Radiation Medicine, University of Kentucky, Lexington, Kentucky, USA.
  • Carlos E Vargas
    Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, Arizona.
  • Nathan Y Yu
    Department of Radiation Oncology, Mayo Clinic, Phoenix, AZ (Y.H., J.H., S.H.P., N.Y.Y., W.L.); Cornell University, Ithaca, NY (Y.H.); Department of Electric Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA (Y.H.); and Department of Radiation Oncology, Mayo Clinic, Rochester, MN (A.B., E.L.M., D.K.E., D.M.R., S.S., C.L.H., B.E.B., M.W.).
  • Brady S Laughlin
    Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, Arizona, USA.
  • Diego Santos Toesca
    Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, Arizona, USA.
  • Sameer Keole
    Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, Arizona, USA.
  • Jean Claude M Rwigema
    Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, Arizona, USA.
  • William W Wong
    Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, Arizona.
  • Steven E Schild
    Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, Arizona.
  • Xue Feng
    Department of Radiology and Medical Imaging, University of Virginia, Charlottesville, Virginia.
  • Quan Chen
    Management School, Zhongshan Institute, University of Electronic Science and Technology of China, Guangdong, 528402, China.
  • Yi Rong
    Department of Radiation Oncology, University of California Davis Medical Center, Sacramento, CA, United States.