Skip to main navigation Skip to search Skip to main content

Anatomy-Guided Synthesis of Novel CT Images at Full Hounsfield Range

  • Stony Brook University
  • Rensselaer Polytechnic Institute

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this work, we present an approach that can synthesize novel CT images across the full Hounsfield range using a very small annotated dataset of around thirty patients and a large non-annotated dataset with high resolution medical images. Our method uses these two datasets in a sequence of steps involving texture learning via StyleGAN and semi-supervised learning via CycleGAN to generate a large annotated medical dataset suitable for use in deep learning algorithms for medical applications. Using an anatomy exploration interface we can then generate CT images with anatomies that were non-existent within either of the datasets, without compromising accuracy and quality. We show that our approach works for all Hounsfield windows with minimal depreciation in anatomical plausibility.

Original languageEnglish
Title of host publication2022 IEEE NSS/MIC RTSD - IEEE Nuclear Science Symposium, Medical Imaging Conference and Room Temperature Semiconductor Detector Conference
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665488723
DOIs
StatePublished - 2022
Event2022 IEEE Nuclear Science Symposium, Medical Imaging Conference, and Room Temperature Semiconductor Detector Conference, IEEE NSS MIC RTSD 2022 - Milano, Italy
Duration: Nov 5 2022Nov 12 2022

Publication series

Name2022 IEEE NSS/MIC RTSD - IEEE Nuclear Science Symposium, Medical Imaging Conference and Room Temperature Semiconductor Detector Conference

Conference

Conference2022 IEEE Nuclear Science Symposium, Medical Imaging Conference, and Room Temperature Semiconductor Detector Conference, IEEE NSS MIC RTSD 2022
Country/TerritoryItaly
CityMilano
Period11/5/2211/12/22

Keywords

  • deep learning
  • GAN
  • medical imaging
  • style loss

Fingerprint

Dive into the research topics of 'Anatomy-Guided Synthesis of Novel CT Images at Full Hounsfield Range'. Together they form a unique fingerprint.

Cite this