®®®® SIIA Público

Título del libro: 2020 Ieee Nuclear Science Symposium And Medical Imaging Conference, Nss/mic 2020
Título del capítulo: Building Blocks for Deep Learning-Based Motion Correction in PET

Autores UNAM:
FRANCISCO EDUARDO ENRIQUEZ MIER Y TERAN;
Autores externos:
Idioma:
Año de publicación:
2020
Palabras clave:

Convolutional neural networks; Deep learning; Diseases; Medical imaging; Building blockes; Convolutional neural network; Imaging problems; Labeled data; Large amounts; Motion correction; Motion parameters; Rigid motions; Rigid-body motion; Supervised neural networks; Engines

Resumen:

Rigid motion correction remains an important challenge in PET. Convolutional Neural Networks (CNNs) have recently matched or outperformed traditional solutions for some medical imaging problems but are largely unexplored for PET motion correction. A limitation of applying supervised neural networks is the requirement for large amounts of labelled data - and this is either unavailable or impractical for the large rigid-body motion parameter space. Therefore, the development of novel and efficient techniques to robustly train CNN models for this application is crucial. Here we present a potential PET ?data engine? that efficiently generates realistic motion-corrupted exemplars. We investigated the feasibility of the engine using simulated data of an [18F]FDG brain PET exam and examined factors that could potentially affect the accuracy of the engine's exemplars, such as the reconstruction algorithms and attenuation correction. © 2020 IEEE

Entidades citadas de la UNAM:
Fuente:


ISBN:
9781728176932
Editorial:
Institute of Electrical and Electronics Engineers Inc.

Tipo de producto:
Conference Paper



DOI:
10.1109/NSS/MIC42677.2020.9508099