Automated in-silico Phantoms for Atherosclerotic Plaque Classification Models from Spectral CT

Goldmann F, Goldmann M, Wels M, Allmendinger T, Sühling M, Maier A (2024)

Publication Type: Conference contribution

Publication year: 2024

Publisher: IEEE

Pages Range: 1-1

Conference Proceedings Title: 2024 IEEE Nuclear Science Symposium (NSS), Medical Imaging Conference (MIC) and Room Temperature Semiconductor Detector Conference (RTSD)

Event location: Tampa US

DOI: 10.1109/NSS/MIC/RTSD57108.2024.10656072

Abstract

We present a novel approach to study atherosclerotic plaque classification models in spectral computed tomography. We propose an end-to-end pipeline for automatically generated, realistic synthetic phantoms of coronary and carotid plaques. Motivated by the need for a more precise coronary artery disease diagnosis, we aim to overcome current limitations such as inaccurate tissue quantification capabilities by conventional methods and insufficient real-world data availability for model validation. The phantoms from our pipeline are based on clinically validated parameters and material compositions to mimic real anatomical conditions. Our approach leverages a state-of-the-art, physics-based ray-tracing framework for simulated data acquisition, which is then processed using calibrated reconstruction software to ensure highly realistic simulated imaging. We conduct an experimental validation consisting of qualitative and quantitative analyses. Comparing signal-to-noise ratio and contrast-to-noise ratio shows similar values for both synthetic and real-world data. The clinical impact of this work is relevant, as it may enable the bootstrapping of deep learning models for plaque classification, potentially enhancing diagnostic accuracy and allow more differentiated treatment strategies.

Authors with CRIS profile

Florian Goldmann Lehrstuhl für Informatik 5 (Mustererkennung) Manuela Goldmann Lehrstuhl für Informatik 5 (Mustererkennung)

Involved external institutions

Siemens Healthineers DE Germany (DE) Siemens AG, Healthcare Sector DE Germany (DE)

How to cite

APA:

Goldmann, F., Goldmann, M., Wels, M., Allmendinger, T., Sühling, M., & Maier, A. (2024). Automated in-silico Phantoms for Atherosclerotic Plaque Classification Models from Spectral CT. In 2024 IEEE Nuclear Science Symposium (NSS), Medical Imaging Conference (MIC) and Room Temperature Semiconductor Detector Conference (RTSD) (pp. 1-1). Tampa, US: IEEE.

MLA:

Goldmann, Florian, et al. "Automated in-silico Phantoms for Atherosclerotic Plaque Classification Models from Spectral CT." Proceedings of the 2024 IEEE Nuclear Science Symposium (NSS), Medical Imaging Conference (MIC) and Room Temperature Semiconductor Detector Conference (RTSD), Tampa IEEE, 2024. 1-1.

BibTeX: Download