Accelerating Radiation Therapy Dose Calculation with Nvidia GPUs

TL;DR

This paper demonstrates that using Nvidia GPUs with a custom Sparse Matrix-Vector Multiplication implementation significantly accelerates radiation therapy dose calculations, reducing computation time and enabling more advanced treatment planning.

Contribution

The paper introduces a highly optimized GPU-based SpMV implementation tailored for RTP, outperforming existing libraries and reducing dose calculation times.

Findings

GPU implementation is up to 4 times faster than baseline.

Custom SpMV has higher operational intensity than existing libraries.

Accelerated dose calculation enables more sophisticated treatment planning.

Abstract

Radiation Treatment Planning (RTP) is the process of planning the appropriate external beam radiotherapy to combat cancer in human patients. RTP is a complex and compute-intensive task, which often takes a long time (several hours) to compute. Reducing this time allows for higher productivity at clinics and more sophisticated treatment planning, which can materialize in better treatments. The state-of-the-art in medical facilities uses general-purpose processors (CPUs) to perform many steps in the RTP process. In this paper, we explore the use of accelerators to reduce RTP calculating time. We focus on the step that calculates the dose using the Graphics Processing Unit (GPU), which we believe is an excellent candidate for this computation type. Next, we create a highly optimized implementation for a custom Sparse Matrix-Vector Multiplication (SpMV) that operates on numerical formats unavailable in state-of-the-art SpMV libraries (e.g., Ginkgo and cuSPARSE). We show that our implementation is several times faster than the baseline (up-to 4x) and has a higher operational intensity than similar (but different) versions such as Ginkgo and cuSPARSE.