EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PocketNet: A Smaller Neural Network for Medical Image Analysis

Adrian Celaya, Jonas A Actor, Rajarajesawari Muthusivarajan, Evan Gates, Caroline Chung, Dawid Schellingerhout, Beatrice Riviere, David Fuentes

IEEE Transactions on Medical Imaging · 2023-04 · 42 (4) · 1172-1184

DOI: 10.1109/tmi.2022.3224873 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2023-04. Verfasst von Adrian Celaya und 7 weiteren Autor:innen. Hauptthemen laut MEDLINE: Neural Networks, Computer, Diagnostic Imaging. Die Arbeit wurde bisher 41-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NLM NIH HHS, Tumor Measurement Initiative through the MD Anderson Strategic Research Initiative Development.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Medical imaging deep learning models are often large and complex, requiring specialized hardware to train and evaluate these models. To address such issues, we propose the PocketNet paradigm to reduce the size of deep learning models by throttling the growth of the number of channels in convolutional neural networks. We demonstrate that, for a range of segmentation and classification tasks, PocketNet architectures produce results comparable to that of conventional neural networks while reducing the number of parameters by multiple orders of magnitude, using up to 90% less GPU memory, and speeding up training times by up to 40%, thereby allowing such models to be trained and deployed in resource-constrained settings.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerDiagnostic Imaging

Förderung

  • NCI NIH HHS
  • NLM NIH HHS
  • Tumor Measurement Initiative through the MD Anderson Strategic Research Initiative Development
  • National Science Foundation
  • Department of Defense through the National Defense Science and Engineering Graduate Fellowship Program
  • Training Fellowship from the Gulf Coast Consortia, on NLM Training Program in Biomedical Informatics and Data Science
  • Dawid Schellingerhout and David Fuentes
  • Beatrice Riviere

Zitieren

APA

Celaya, A., Actor, J. A., Muthusivarajan, R., Gates, E., Chung, C., Schellingerhout, D., Riviere, B., & Fuentes, D. (2023). PocketNet: A Smaller Neural Network for Medical Image Analysis. IEEE Transactions on Medical Imaging, 42(4), 1172-1184. https://doi.org/10.1109/tmi.2022.3224873

Vancouver

Celaya A, Actor JA, Muthusivarajan R, Gates E, Chung C, Schellingerhout D, et al. PocketNet: A Smaller Neural Network for Medical Image Analysis. IEEE Transactions on Medical Imaging. 2023;42(4):1172-1184. doi:10.1109/tmi.2022.3224873

BibTeX
@article{Celaya2023,
  author = {Celaya, Adrian and Actor, Jonas A and Muthusivarajan, Rajarajesawari and Gates, Evan and Chung, Caroline and Schellingerhout, Dawid and Riviere, Beatrice and Fuentes, David},
  title = {PocketNet: A Smaller Neural Network for Medical Image Analysis},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2023},
  volume = {42},
  number = {4},
  pages = {1172--1184},
  doi = {10.1109/tmi.2022.3224873}
}
Kostenlos starten →