Efficient 3D CT scanning on small computers beats larger machines
An Efficient Out-of-Core Tomographic Imaging Framework for Edge Devices
Distributed, Parallel, and Cluster Computing
Summary
CT scanning creates 3D pictures used in medicine, but it takes a lot of computer power and memory, which small devices often lack. The researchers designed a new method, called edgeFBP, that helps small devices like Nvidia Jetson work faster and use less energy when making these 3D images. They improved the process by using special computer chips and smart shortcuts that speed up the hardest parts of the calculation. Their system runs much more efficiently than existing software on small devices and uses much less power compared to big, powerful machines used in data centers. This could make advanced 3D scanning more accessible on portable and low-power devices.
Computed tomographyCT reconstructionEdge devicesNvidia JetsonMixed-precision computingTensor CoresBack-projectionEnergy efficiencyOut-of-core computingSystem-on-Chip
Authors
Xuetao Chen, Cong Ma, Xiangyu Meng, Du Wu, Zhengyang Bai, Tao Luo, Zhaorui Zhang, Emmanuel Jeannot, Edgar Josafat Martinez Noriega, Xun Wang, Peng Chen, Amelie Chi Zhou, Mohamed Wahib
Abstract
Computed Tomography (CT) is an essential 3D imaging technology widely used in medical diagnostics and scientific research. However, performing CT imaging on edge devices is challenging due to limitations in computational power, memory capacity, and energy budget. This paper presents an efficient CT reconstruction framework, called edgeFBP, designed for Nvidia Jetson System-on-Chip (SoC) devices. edgeFBP adopts an end-to-end pipeline design for efficient out-of-core image reconstruction under tight power and memory constraints. edgeFBP utilizes a mixed-precision strategy leveraging half-precision Tensor Cores (TCs) to accelerate the bottleneck back-projection (BP) kernel. edgeFBP achieves a 1.83x speedup over the widely used RTK library on Jetson Nano and a 2.56x speedup on Jetson AGX. Under a strict 25-Watt power budget, edgeFBP on Jetson Nano achieves up to 5-48x higher energy efficiency than an Nvidia DGX A100, enabling datacenter-scale imaging on constrained edge devices.