Your browser doesn't support javascript.
loading
Student becomes teacher: training faster deep learning lightweight networks for automated identification of optical coherence tomography B-scans of interest using a student-teacher framework.
Owen, Julia P; Blazes, Marian; Manivannan, Niranchana; Lee, Gary C; Yu, Sophia; Durbin, Mary K; Nair, Aditya; Singh, Rishi P; Talcott, Katherine E; Melo, Alline G; Greenlee, Tyler; Chen, Eric R; Conti, Thais F; Lee, Cecilia S; Lee, Aaron Y.
Afiliación
  • Owen JP; Department of Ophthalmology, University of Washington, Seattle, WA 98195, USA.
  • Blazes M; Department of Ophthalmology, University of Washington, Seattle, WA 98195, USA.
  • Manivannan N; Carl Zeiss Meditec, Inc., Dublin, CA 94568, USA.
  • Lee GC; Carl Zeiss Meditec, Inc., Dublin, CA 94568, USA.
  • Yu S; Carl Zeiss Meditec, Inc., Dublin, CA 94568, USA.
  • Durbin MK; Carl Zeiss Meditec, Inc., Dublin, CA 94568, USA.
  • Nair A; Carl Zeiss Meditec, Inc., Dublin, CA 94568, USA.
  • Singh RP; Center for Ophthalmic Bioinformatics, Cole Eye Institute, Cleveland Clinic, Cleveland, OH 44195, USA.
  • Talcott KE; Center for Ophthalmic Bioinformatics, Cole Eye Institute, Cleveland Clinic, Cleveland, OH 44195, USA.
  • Melo AG; Center for Ophthalmic Bioinformatics, Cole Eye Institute, Cleveland Clinic, Cleveland, OH 44195, USA.
  • Greenlee T; Center for Ophthalmic Bioinformatics, Cole Eye Institute, Cleveland Clinic, Cleveland, OH 44195, USA.
  • Chen ER; Center for Ophthalmic Bioinformatics, Cole Eye Institute, Cleveland Clinic, Cleveland, OH 44195, USA.
  • Conti TF; Center for Ophthalmic Bioinformatics, Cole Eye Institute, Cleveland Clinic, Cleveland, OH 44195, USA.
  • Lee CS; Department of Ophthalmology, University of Washington, Seattle, WA 98195, USA.
  • Lee AY; Department of Ophthalmology, University of Washington, Seattle, WA 98195, USA.
Biomed Opt Express ; 12(9): 5387-5399, 2021 Sep 01.
Article en En | MEDLINE | ID: mdl-34692189
This work explores a student-teacher framework that leverages unlabeled images to train lightweight deep learning models with fewer parameters to perform fast automated detection of optical coherence tomography B-scans of interest. Twenty-seven lightweight models (LWMs) from four families of models were trained on expert-labeled B-scans (∼70 K) as either "abnormal" or "normal", which established a baseline performance for the models. Then the LWMs were trained from random initialization using a student-teacher framework to incorporate a large number of unlabeled B-scans (∼500 K). A pre-trained ResNet50 model served as the teacher network. The ResNet50 teacher model achieved 96.0% validation accuracy and the validation accuracy achieved by the LWMs ranged from 89.6% to 95.1%. The best performing LWMs were 2.53 to 4.13 times faster than ResNet50 (0.109s to 0.178s vs. 0.452s). All LWMs benefitted from increasing the training set by including unlabeled B-scans in the student-teacher framework, with several models achieving validation accuracy of 96.0% or higher. The three best-performing models achieved comparable sensitivity and specificity in two hold-out test sets to the teacher network. We demonstrated the effectiveness of a student-teacher framework for training fast LWMs for automated B-scan of interest detection leveraging unlabeled, routinely-available data.
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Tipo de estudio: Diagnostic_studies Idioma: En Revista: Biomed Opt Express Año: 2021 Tipo del documento: Article País de afiliación: Estados Unidos Pais de publicación: Estados Unidos
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Tipo de estudio: Diagnostic_studies Idioma: En Revista: Biomed Opt Express Año: 2021 Tipo del documento: Article País de afiliación: Estados Unidos Pais de publicación: Estados Unidos