Related Experiment Video
Updated: Jan 18, 2026
High-Throughput Total Internal Reflection Fluorescence and Direct Stochastic Optical Reconstruction Microscopy Using a Photonic Chip
Published on: November 16, 2019
Edge computing-based FPGA real-time material decomposition system for photon counting CT
Mengqing Su1, Xiaopeng Yu2, Qianyu Wu1
1Laboratory of Image Science and Technology, School of Computer Science and Engineering, Southeast University, Nanjing, 210096, China; School of Biomedical Engineering, ShanghaiTech University, Shanghai, 201210, China; Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing, Southeast University, Nanjing, 210096, China.
Photon counting computed tomography (PCCT) faces data challenges. This study integrates material decomposition into field-programmable gate arrays (FPGAs) within the CT gantry, enabling faster, high-precision imaging.
Area of Science:
- Medical Imaging
- Computed Tomography
- Data Processing
Background:
- Photon counting computed tomography (PCCT) offers superior spatial resolution using photon counting detectors (PCDs).
- PCCT generates significantly larger raw data (20-100x traditional CT) due to multiple energy bins and smaller pixels.
- Existing slip ring bandwidth limits the transfer of large PCCT datasets for offline processing.
Purpose of the Study:
- To develop an efficient edge computing solution for PCCT data processing.
- To implement real-time material decomposition directly on an FPGA within the CT gantry.
- To overcome data transfer bottlenecks associated with high-resolution PCCT.
Main Methods:
- An edge computing framework was designed for PCCT data processing.
- A fast material decomposition algorithm was developed and implemented on an FPGA.
- The processing workflow was shifted from offline analysis to the CT gantry, utilizing integrated FPGA resources.
Main Results:
- The proposed system successfully generated ring-artifact-free material decomposition results.
- Accurate material decomposition was achieved with reduced data volume.
- Real phantom datasets demonstrated the system's efficiency.
Conclusions:
- Offline material decomposition processing was migrated to the detector blade using FPGA resources.
- The system significantly enhances processing speed and throughput compared to traditional methods.
- The proposed approach maintains the precision of offline processing while enabling rapid results.
