TY - GEN
T1 - Multi-spectral In-Vivo FPGA-Based Surgical Imaging
AU - Alsharari, Majed
AU - Niemitz, Lorenzo
AU - Sorensen, Simon
AU - Woods, Roger
AU - Burke, Ray
AU - Engels, Stefan Andersson
AU - Reaño, Carlos
AU - Mai, Son T.
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - Intelligent and adaptive in-vivo, catheter-based imaging systems with enhanced processing and analytical capability have the potential to enhance surgical operations and improve patient care. The paper describes an intelligent surgical imaging system based on a ‘chip on tip’, which reduces the need for conventional imaging. The associated embedded system provides real-time, in-vivo imaging analysis and data display for surgeons, enhancing their ability to detect clinically significant tissue. The paper presents initial work on an field programmable gate array implementation of a contrast limited adaptive histogram equalization algorithm, Hessian matrix construction and region of interest function on the AMD-Xilinx’s Kria KV260 board. It outlines optimizations undertaken to reduce the BRAMs by 38%, DSP48 blocks by 80%, flip-flops by 33% and LUTs by 36%, thus creating a design operating at 121 FPS.
AB - Intelligent and adaptive in-vivo, catheter-based imaging systems with enhanced processing and analytical capability have the potential to enhance surgical operations and improve patient care. The paper describes an intelligent surgical imaging system based on a ‘chip on tip’, which reduces the need for conventional imaging. The associated embedded system provides real-time, in-vivo imaging analysis and data display for surgeons, enhancing their ability to detect clinically significant tissue. The paper presents initial work on an field programmable gate array implementation of a contrast limited adaptive histogram equalization algorithm, Hessian matrix construction and region of interest function on the AMD-Xilinx’s Kria KV260 board. It outlines optimizations undertaken to reduce the BRAMs by 38%, DSP48 blocks by 80%, flip-flops by 33% and LUTs by 36%, thus creating a design operating at 121 FPS.
KW - Field programmable gate array
KW - Surgical imaging
UR - https://www.scopus.com/pages/publications/85142767136
U2 - 10.1007/978-3-031-19983-7_8
DO - 10.1007/978-3-031-19983-7_8
M3 - Conference proceeding
AN - SCOPUS:85142767136
SN - 9783031199820
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 103
EP - 117
BT - Applied Reconfigurable Computing. Architectures, Tools, and Applications - 18th International Symposium, ARC 2022, Proceedings
A2 - Gan, Lin
A2 - Wang, Yu
A2 - Xue, Wei
A2 - Chau, Thomas
PB - Springer Science and Business Media Deutschland GmbH
T2 - 18th International Symposium on Applied Reconfigurable Computing, ARC 2022
Y2 - 19 September 2022 through 20 September 2022
ER -