TY  - JOUR
AB  - We present a new approach for deep learned joint PET-MR image reconstruction inspired by conventional synergistic methods using a joint regularizer. The maximum a posteriori expectation-maximization algorithm for PET and the Landweber algorithm for MR are unrolled and interconnected through a deep learned joint regularization step. The parameters of the joint U-Net regularizer and the respective regularization strengths are learned and shared across all the iterations. Along with introducing this framework, we propose an investigation of the impact of the loss function selection on network performance. We explored how the network performs when trained with a single or a joint-modality loss. Finally, we explored under which settings a joint reconstruction was beneficial for MR reconstruction by using various undersampling factors. The results obtained on 2-D simulated data show that the joint networks outperform conventional synergistic methods and independent deep learned reconstruction methods. For PET, the network trained with only a PET loss achieves a better global reconstruction accuracy than the version trained with a weighted sum of PET and MR loss terms. More importantly, the former further improves the reconstruction of PET-specific features where MR-guided methods show their limit. Therefore, using a single-modality loss to supervise the training while still reconstructing the two modalities in parallel leads to better reconstructions and improved modality-unique lesion recovery in our proposed framework. For MR, while the same effect is observed, joint reconstruction gains only occur in the presence of highly undersampled data. Single-modality loss joint reconstruction results are also demonstrated on 3-D clinical PET-MR datasets.
AU  - Corda-D'Incan, G.*
AU  - Schnabel, J.A.
AU  - Hammers, A.*
AU  - Reader, A.J.*
C1  - 68533
C2  - 53665
CY  - 445 Hoes Lane, Piscataway, Nj 08855-4141 Usa
SP  - 742-754
TI  - Single-modality supervised joint PET-MR image reconstruction.
JO  - IEEE TRPMS
VL  - 7
IS  - 7
PB  - Ieee-inst Electrical Electronics Engineers Inc
PY  - 2023
SN  - 2469-7311
ER  - 

TY  - JOUR
AB  - Over the past decades, significant improvements have been made in the field of computational human phantoms (CHPs) and their applications in biomedical engineering. Their sophistication has dramatically increased. The very first CHPs were composed of simple geometric volumes, e.g., cylinders and spheres, while current CHPs have a high resolution, cover a substantial range of the patient population, have high anatomical accuracy, are poseable, morphable, and are augmented with various details to perform functionalized computations. Advances in imaging techniques and semi-automated segmentation tools allow fast and personalized development of CHPs. These advances open the door to quickly develop personalized CHPs, inherently including the disease of the patient. Because many of these CHPs are increasingly providing data for regulatory submissions of various medical devices, the validity, anatomical accuracy, and availability to cover the entire patient population is of utmost importance. The article is organized into two main sections: the first section reviews the different modeling techniques used to create CHPs, whereas the second section discusses various applications of CHPs in biomedical engineering. Each topic gives an overview, a brief history, recent developments, and an outlook into the future.
AU  - Kainz, W.*
AU  - Neufeld, E.*
AU  - Bolch, W.E.*
AU  - Graff, C.G.*
AU  - Kim, C.H.*
AU  - Kuster, N.*
AU  - Lloyd, B.*
AU  - Morrison, T.*
AU  - Segars, P.*
AU  - Yeom, Y.S.*
AU  - Zankl, M.
AU  - Xu, X.G.*
AU  - Tsui, B.M.W.*
C1  - 55427
C2  - 46115
SP  - 1-23
TI  - Advances in computational human phantoms and their applications in biomedical engineering - A topical review.
JO  - IEEE TRPMS
VL  - 3
IS  - 1
PY  - 2019
SN  - 2469-7311
ER  - 

TY  - JOUR
AB  - The aim of the paper is to explore a new method for organ contour description in radiology and radiation protection. The method bases on the mathematical computation of electrical fields, exploited are the equipotential lines caused by a potential field of a distribution of point sources in analogy to electric charges. The organ shape is described by the potential values of the field, the contour by the equipotentials. The potential-dependent methods offers an inside-outside criterion and can be scaled in size and edited by changing the source points. Because of that it offers a flexible possible framework for organ contour editing and also towards segmentation. The main focus of the paper is the proof of principle, i.e. the optimization of the source point coordinates and source strengths, to show the transfer of voxelized organ borders to potential based contours. The already voxelized organ borders were from a human voxel phantom generated from 2-dimensional CT images of a real patient. Results for several closed and compact organs shall be presented and the limitations, future applications and possibilities addressed, e.g. the advantages of an implementation in Monte Carlo calculations of radiation transport.
AU  - Becker, J.
AU  - Fedrigo, M.*
C1  - 54671
C2  - 45749
SP  - 24-30
TI  - Introducing the concept of potential-based organ contours.
JO  - IEEE TRPMS
VL  - 3
IS  - 1
PY  - 2018
SN  - 2469-7311
ER  -