Accelerating Fourier Domain Optical Coherence Tomography using general purpose graphics processing units and field programmable gate arrays

摘要

Fourier Domain Optical Coherence Tomography is an emerging biomedical imaging technology that provides ultra high resolution and a fast imaging speed. The complexity of the FD-OCT algorithm demands high processing power from the underlying platform. However, the scaling of faster data acquisition rates and 3D imaging on real time FD-OCT systems is quickly outpacing the performance growth of General Purpose Processors. Our research investigates the scalability of two potential platforms for accelerating FD-OCT — GPGPUs and FPGAs. We implemented a FD-OCT system using a GPGPU as co-processor with 6.9x speed up. We also created a hardware processing engine using FPGAs, delivering over 2x the throughput rate over GPGPU with 1024- point FFT. Our analysis on the performance and scalability for both platforms shows that, while GPGPUs offer an easy and low cost solution for accelerating FD-OCT, FPGAs are more likely to match the long term demands for real-time, 3D FD-OCT.