Skip to content

Announcement of a Defense of a Dissertation- William Whiteley

Candidate for Doctor of Philosophy
Defense of a Ph.D. Dissertation
Faculty Advisor: Jens Gregor
Where: See Zoom information below

Deep Learning in Positron Emission Tomography Image Reconstruction

The aim of this research is towards creating superior algorithms for positron emission tomography (PET) image reconstruction through the application of deep learning methods. The central idea is applying the increasing availability of computational power and deep neural network techniques to develop superior image reconstruction algorithms and better health outcomes for patients. This dissertation is structured as a series of four articles detailing deep learning methods applied to the PET image formation process. The first article proposes a raw data correction method for the case where one or more block detectors in the PET scanner fail. Gaps in the raw data are repaired with a neural network to allow the continued scanning of patients with full diagnostic quality until a hardware repair can be made.  The remaining articles explore reconstruction methods where a neural network is the primary instrument for generating an image directly from measurement data.  In the second article, a method named DirectPET is described that works directly from raw sinograms and attenuation maps to create multi-slice full-size image volumes with a purposely designed Radon inversion layer. DirectPET generates images that are quantitatively and qualitatively similar to the conventional ordered subset expectation maximization (OSEM) method in a fraction of the reconstruction time. Continuing on the theme of direct neural network reconstruction, the next article proposes a network named FastPET that works from histo-image data and is capable of creating high quality PET images over 800 times faster than conventional methods while demonstrating a higher contrast recovery than OSEM with equivalent variance.  The final article explores a network architecture that is a hybrid between DirectPET and FastPET. It utilizes sinograms as input and the fully-connected Radon inversion layer like DirectPET, but also uses histo-images and encoder-decoder elements like FastPET. This multimodal architecture named MDPET is shown to provide the highest overall image quality of any known direct neural network PET reconstruction method.

Dial-In Information

Zoom information:

William James Whiteley is inviting you to a scheduled Zoom meeting.
Topic: Deep Learning in Positron Emission Tomography
Time: Mar 27, 2020 01:00 PM Eastern Time (US and Canada)

Join from PC, Mac, Linux, iOS or Android:

Or iPhone one-tap (US Toll):  +13126266799,979517062#  or +16468769923,979517062#

Or Telephone:
    +1 312 626 6799 (US Toll)
    +1 646 876 9923 (US Toll)
    +1 346 248 7799 (US Toll)
    +1 669 900 6833 (US Toll)
    +1 253 215 8782 (US Toll)
    +1 301 715 8592 (US Toll)
    Meeting ID: 979 517 062

International numbers available:

Friday, March 27, 2020 at 1:00pm to 3:00pm

Virtual Event

Recent Activity