Case Study: M&R Technology accelerates medical ray-tracing simulations with MathWorks GPU Coder

M&R Technology Accelerates Medical Ray-Tracing Algorithms on Desktop Computers Using GPU Coder and Automatic CUDA Code Generation

M&R Technology, a San Diego-based developer of sensor solutions for next-generation IoT applications, needed to speed up a desktop application that simulates medical photoplethysmography (PPG) signals using ray tracing in a tissue model. Their CPU-based computations were substantial and could take several hours, especially for large simulation datasets. MathWorks provided GPU Coder to help address this performance challenge.

Using MathWorks GPU Coder and automatic CUDA code generation, M&R Technology generated custom GPU-accelerated algorithms and leveraged the cuBLAS library to optimize key operations such as 3D distance calculations, trigonometric matrix computations, and matrix multiplications. The result was a dramatic speedup—computation times dropped from hours to minutes or even seconds, with performance improvements of several hundred times depending on data size.

M&R Technology

Jerker Taudien

M&R Technology