Accurate and efficient Methods for Fluorescence-mediated Tomography

• Applicant: Privatdozent Dr. Felix Gremse
• Subject Area: Medical Physics, Biomedical Technology ; Epidemiology and Medical Biometry/Statistics ; Computer Architecture, Embedded and Massively Parallel Systems
• Term: from 2017 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 380646556

• Final Report Year: 2023

Final Report Abstract

Multimodal fluorescence tomography (FMT) allows noninvasive quantitative determination of three-dimensional fluorescence distribution in small objects, such as body parts or laboratory mice. However, fluorescence tomography is a mathematically and numerically challenging inverse problem, which requires accurate and efficient calculation of the light distributions in the object and complex reconstruction. The aim of this project was to perform Monte Carlo simulations of the light behavior in the near-infrared range in mouse bodies to generate artificial raw data to determine factors influencing the reconstruction quality. Furthermore, numerical methods such as implicit matrix representations and GPU-accelerated multigrid methods were to be investigated. The goals also included an experimental investigation using multimodal imaging to evaluate and exploit improvements in fluorescence reconstruction. In this project, we developed GPU-accelerated Monte Carlo simulations to compute artificial raw data. Since hundreds of simulations are required per scan, we investigated variance reduction techniques, dynamically splitting and merging photons to invest more and less effort in unknown and already known regions, respectively. Furthermore, we simulated the background signal, which has great practical relevance but also, together with variance reduction, provides an acceleration of the computation time by several orders of magnitude. To reduce memory requirements for the reconstruction matrix, an implicit matrix representation was implemented, which allows to use all measurement data for the reconstruction and thus to achieve a higher resolution. For this purpose, suitable iterative non-negative linear solvers and possibilities for convergence acceleration were investigated and integrated. Furthermore, reconstruction possibilities for the determination of absorption and scattering maps were investigated, where we can at least determine the values in homogenized organs. The improvements in fluorescence reconstruction were experimentally validated with in vivo PET-CT and CT-FMT experiments, using a double-labeled sample (fluorescent and radioactive). Furthermore, we support colleagues and collaborators in their CT-FMT experiments and could contribute to several publications.

Publications

• A preclinical micro-computed tomography database including 3D whole body organ segmentations. Scientific Data, 5(1).
  Rosenhain, Stefanie; Magnuska, Zuzanna A.; Yamoah, Grace G.; Al Rawashdeh, Wa’el; Kiessling, Fabian & Gremse, Felix
  
• Histidine-rich glycoprotein-induced vascular normalization improves EPR-mediated drug targeting to and into tumors. Journal of Controlled Release, 282, 25-34.
  Theek, Benjamin; Baues, Maike; Gremse, Felix; Pola, Robert; Pechar, Michal; Negwer, Inka; Koynov, Kaloian; Weber, Benjamin; Barz, Matthias; Jahnen-Dechent, Willi; Storm, Gert; Kiessling, Fabian & Lammers, Twan
  
• Memory-Efficient Sparse Matrix-Matrix Multiplication by Row Merging on Many-Core Architectures. SIAM Journal on Scientific Computing, 40(4), C429-C449.
  Gremse, Felix; Küpper, Kerstin & Naumann, Uwe
  
• Data Curation for Preclinical and Clinical Multimodal Imaging Studies. Molecular Imaging and Biology, 21(6), 1034-1043.
  Yamoah, Grace Gyamfuah; Cao, Liji; Wu, Chao Wu; Beekman, Freek J.; Vandeghinste, Bert; Mannheim, Julia G.; Rosenhain, Stefanie; Leonardic, Kevin; Kiessling, Fabian & Gremse, Felix
  
• Non-invasive Imaging and Modeling of Liver Regeneration After Partial Hepatectomy. Frontiers in Physiology, 10.
  Zafarnia, Sara; Mrugalla, Anna; Rix, Anne; Doleschel, Dennis; Gremse, Felix; Wolf, Stephanie D.; Buyel, Johannes F.; Albrecht, Ute; Bode, Johannes G.; Kiessling, Fabian & Lederle, Wiltrud
  
• A collagen-binding protein enables molecular imaging of kidney fibrosis in vivo. Kidney International, 97(3), 609-614.
  Baues, Maike; Klinkhammer, Barbara M.; Ehling, Josef; Gremse, Felix; van Zandvoort Marc, A.M.J.; Reutelingsperger, Chris P.M.; Daniel, Christoph; Amann, Kerstin; Bábíčková, Janka; Kiessling, Fabian; Floege, Jürgen; Lammers, Twan & Boor, Peter
  
• Multimodal and multiscale optical imaging of nanomedicine delivery across the blood-brain barrier upon sonopermeation. Theranostics, 10(4), 1948–1959.
  May, Jan-Niklas; Golombek, Susanne K.; Baues, Maike; Dasgupta, Anshuman; Drude, Natascha; Rix, Anne; Rommel, Dirk; von Stillfried, Saskia; Appold, Lia; Pola, Robert; Pechar, Michal; van Bloois, Louis; Storm, Gert; Kuehne, Alexander J. C.; Gremse, Felix; Theek, Benjamin; Kiessling, Fabian & Lammers, Twan
  
• Mixing Matrix-corrected Whole-body Pharmacokinetic Modeling Using Longitudinal Micro-computed Tomography and Fluorescence-mediated Tomography. Molecular Imaging and Biology, 23(6), 963-974.
  Zuo, Simin; Al Rawashdeh, Wa’el; Rosenhain, Stefanie; Magnuska, Zuzanna; Gyamfuah, Yamoah Grace; Kiessling, Fabian & Gremse, Felix
  
• A Micro-computed Tomography Database and Reference Implementation for Parallel Computations of Trabecular Thickness and Spacing. Journal of Open Research Software, 10(1), 4.
  Nguyen, Thi-Ngoc-Thu; Höfter, Andreas; Leonardic, Kevin; Rosenhain, Stefanie; Kiessling, Fabian; Sae-Tang, Wannida; Naumann, Uwe & Gremse, Felix
  
• Intrinsic Respiratory Gating for Simultaneous Multi-Mouse μCT Imaging to Assess Liver Tumors. Frontiers in Medicine, 9.
  Thamm, Mirko; Rosenhain, Stefanie; Leonardic, Kevin; Höfter, Andreas; Kiessling, Fabian; Osl, Franz; Pöschinger, Thomas & Gremse, Felix