Learning efficient channels with a dual loss autoencoder

Jason L. Granstedt, Weimin Zhou, Mark A. Anastasio

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In medical imaging systems, task-based metrics have been advocated as a means of evaluating image quality. Mathematical observers are one method of computing such metrics. Although the Bayesian Ideal Observer (IO) is optimal by definition, it is frequently intractable and non-linear. Linear approximations to the IO are sometimes employed to obtain task-based statistics when computing the IO is infeasible. The optimal linear observer for maximizing the SNR of the test statistic is the Hotelling Observer (HO). However, the computational cost for computing the HO increases with image size and becomes intractable for larger images. Channelized methods of reducing the dimensionality of the data before computing the HO have become popular, with efficient channels capable of approximating the HO's performance at significantly reduced computational cost. State-of-the-art channels have been learned by using an autoencoder (AE) to encode data by employing a known signal template as the desired reconstruction, but the method is dependant on a high-quality estimate of the signal. An alternative to channels is approximating the test statistic directly using a feed-forward neural network (FFNN). However, this approach can overfit when the amount of training data is limited. In this work, a generalized method for learning channels utilizing an AE with dual losses (AEDL) is proposed. The AEDL framework jointly minimizes both task-specific and reconstruction losses to learn a set of efficient channels, even when the number of training images is relatively small. Preliminary results indicate that the proposed network outperforms state-of-the-art methods on the selected imaging task. Additionally, the AEDL framework suffers from less overfitting than the FFNN.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2020
Subtitle of host publicationImage Perception, Observer Performance, and Technology Assessment
EditorsFrank W. Samuelson, Sian Taylor-Phillips
PublisherSPIE
ISBN (Electronic)9781510633995
DOIs
StatePublished - Jan 1 2020
EventMedical Imaging 2020: Image Perception, Observer Performance, and Technology Assessment - Houston, United States
Duration: Feb 19 2020Feb 20 2020

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume11316
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2020: Image Perception, Observer Performance, and Technology Assessment
CountryUnited States
CityHouston
Period2/19/202/20/20

Keywords

  • Autoencoders
  • Imaging system optimization
  • Neural networks
  • Numerical observers
  • Objective assessment of image quality

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Learning efficient channels with a dual loss autoencoder'. Together they form a unique fingerprint.

  • Cite this

    Granstedt, J. L., Zhou, W., & Anastasio, M. A. (2020). Learning efficient channels with a dual loss autoencoder. In F. W. Samuelson, & S. Taylor-Phillips (Eds.), Medical Imaging 2020: Image Perception, Observer Performance, and Technology Assessment [113160C] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 11316). SPIE. https://doi.org/10.1117/12.2549363