Exploiting Heterogeneous Architectures for Rigid Image Registration

D'Arnese, Eleonora; Sozzo, Emanuele Del; Conficconi, Davide; Santambrogio, Marco D.

doi:10.1109/biocas49922.2021.9645026

Cited by 4 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, none of these algorithms achieves perfect accuracy since it would be unfeasible in terms of execution time. Consequently, other solutions exploit hardware acceleration to enable more accurate algorithms while keeping low execution time [16], [24]- [26]. On this path, GPUs proved very effective in accelerating IR algorithms.…”

Section: Related Workmentioning

confidence: 99%

“…On this path, GPUs proved very effective in accelerating IR algorithms. While many solutions performed approximations to achieve significant results [26], D'Arnese et al [16] developed a framework to accelerate 2D IR by employing heterogeneous architecture without reducing accuracy and reaching remarkable performance. However, they only focus on 2D IR, which does not take advantage of the information about the overall 3D volume.…”

Section: Related Workmentioning

confidence: 99%

“…In this sense, recognizing the effort of the approach proposed in [16], we present ATHENA, a framework exploiting heterogeneous architectures to accelerate multi-modal 3D rigid image registration, reaching state-of-the-art accuracy and significantly reducing execution time compared to the software. Differently from others, ATHENA uses information about the volume to register highly misaligned images, thus fixing problems of existing solutions in the state-of-the-art.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

ATHENA: a GPU-based Framework for Biomedical 3D Rigid Image Registration

Sorrentino,

Venere,

D’Arnese

et al. 2023

2023 IEEE Biomedical Circuits and Systems Conference (BioCAS)

Self Cite

View full text Add to dashboard Cite

3D image registration is one of the most complex algorithms employed in medical imaging applications. Software solution struggles to reach high accuracy in a reasonable time, therefore this work presents ATHENA, a framework for rigid 3D Image Registration, exploiting heterogeneous architecture for acceleration, and also providing support for memory-constrained devices. Moreover, ATHENA presents a tool for automatically generating misalignent between volumes, to perform a robustness analysis on different kinds of distortions. We compared ATHENA with SimpleITK, a well-known software library, and with a software version of the proposed algorithm, achieving a top speedup of 18.1× and 53.7× respectively.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

ATHENA: a GPU-based Framework for Biomedical 3D Rigid Image Registration

Sorrentino,

Venere,

D’Arnese

et al. 2023

2023 IEEE Biomedical Circuits and Systems Conference (BioCAS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Nevertheless, it requires programming knowledge and a deep understanding of the algorithm to tune the hyperparameters. On the other hand, hardware (HW) accelerators are gaining traction as an alternative for higher performance and energy efficiency with two main approaches: GPUbased and FPGA-based [3], [7], [8], [9], [10], [11], [12]. The former is a valuable solution when images, particularly volumes, are involved, but it lacks energy efficiency.…”

Section: Introductionmentioning

confidence: 99%

Faber: A Hardware/SoftWare Toolchain for Image Registration

D'Arnese

Conficconi

Sozzo

et al. 2023

IEEE Trans. Parallel Distrib. Syst.

Self Cite

View full text Add to dashboard Cite

Image registration is a well-defined computation paradigm widely applied to align one or more images to a target image. This paradigm, which builds upon three main components, is particularly compute-intensive and represents many image processing pipelines' bottlenecks. State-of-the-art solutions leverage hardware acceleration to speed up image registration, but they are usually limited to implementing a single component. We present Faber, an open-source HW/SW CAD toolchain tailored to image registration. The Faber toolchain comprises HW/SW highly-tunable registration components, supports users with different expertise in building custom pipelines, and automates the design process. In this direction, Faber provides both default settings for entry-level users and latency and resource models to guide HW experts in customizing the different components. Finally, Faber achieves from 1.5× to 54× in speedup and from 2× to 177× in energy efficiency against state-of-the-art tools on a Xeon Gold.

show abstract