Karl Krueger
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Karl Krueger has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Bjoern Weissler
Department of Physics of Molecular Imaging Systems, Institute for Experimental Molecular Imaging, RWTH Aachen University, Aachen, Germany
1 shared publication - David Schug
Department of Physics of Molecular Imaging systems Institute for Experimental Molecular Imaging RWTH Aachen University Aachen Germany
1 shared publication - Florian Mueller
Department of Physics of Molecular Imaging systems Institute for Experimental Molecular Imaging RWTH Aachen University Aachen Germany
1 shared publication - Pierre Gebhardt
Department of Physics of Molecular Imaging Systems, Institute for Experimental Molecular Imaging, RWTH Aachen University, Aachen, Germany
1 shared publication Volkmar Schulz
Department of Physics of Molecular Imaging Systems, Institute for Experimental Molecular Imaging, and the Physics Institute III B, RWTH Aachen University, Aachen, Germany
1 shared publication
Latest Publications
- Open Access
- 23 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Radiation and Plasma Medical Sciences
- Vol. 7 (3) , 253-262
High-Throughput FPGA-Based Inference of Gradient Tree Boosting Models for Position Estimation in PET Detectors
In modern large-scale PET systems, transferring the digitized raw detector data at high count rates to a centralized processing unit is a challenge. Processing data on FPGAs close to the detectors can reduce data early on and improve scalability of the PET system. We present and evaluate an FPGA implementation of gradient tree boosting (GTB) for one-dimensional position estimation of gamma interactions in the scintillator. GTB is a supervised machine learning algorithm based on building ensembles of binary decision trees. Models were trained offline and inferred in an FPGA (XC7K410T-2FFG676 Kintex-7). Input features and GTB parameters influencing both positioning performance and model size were varied while evaluating the inferred models concerning data throughput and FPGA resource consumption as well as positioning performance. We achieved throughputs per detector between 2.94 × 106 and 4.55 × 106 gamma interactions per second. For an optimized GTB model, resource consumption could be reduced by factors of 17 and 10 to less than 1% (2.51 × 103 look-up tables) of available logic and 1.26% (20 BRAMs) of memory resources, while maintaining a positioning performance of 98.63% when compared to the model with the best positioning performance. The presented framework can be easily adapted to other photosensors and scintillator configurations.
- 23 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Radiation and Plasma Medical Sciences
- Vol. 7 (3) , 253-262
Open AccessHigh-Throughput FPGA-Based Inference of Gradient Tree Boosting Models for Position Estimation in PET Detectors
In modern large-scale PET systems, transferring the digitized raw detector data at high count rates to a centralized processing unit is a challenge. Processing data on FPGAs close to the detectors can reduce data early on and improve scalability of the PET system. We present and evaluate an FPGA implementation of gradient tree boosting (GTB) for one-dimensional position estimation of gamma interactions in the scintillator. GTB is a supervised machine learning algorithm based on building ensembles of binary decision trees. Models were trained offline and inferred in an FPGA (XC7K410T-2FFG676 Kintex-7). Input features and GTB parameters influencing both positioning performance and model size were varied while evaluating the inferred models concerning data throughput and FPGA resource consumption as well as positioning performance. We achieved throughputs per detector between 2.94 × 106 and 4.55 × 106 gamma interactions per second. For an optimized GTB model, resource consumption could be reduced by factors of 17 and 10 to less than 1% (2.51 × 103 look-up tables) of available logic and 1.26% (20 BRAMs) of memory resources, while maintaining a positioning performance of 98.63% when compared to the model with the best positioning performance. The presented framework can be easily adapted to other photosensors and scintillator configurations.