Highly Efficient and Scalable Framework for High-Speed Super-Resolution Microscopy

Abstract

The multiple signal classification algorithm (MUSICAL) is a statistical super-resolution technique for wide-field fluorescence microscopy. Although MUSICAL has several advantages, such as its high resolution, its low computational performance has limited its exploitation. This paper aims to analyze the performance and scalability of MUSICAL for improving its low computational performance. We first optimize MUSICAL for performance analysis by using the latest high-performance computing libraries and parallel programming techniques. Afterward, we provide insights into MUSICAL's performance bottlenecks. Based on the insights, we develop a new parallel MUSICAL in C++ using Intel Threading Building Blocks and the Intel Math Kernel Library. Our experimental results show that our new parallel MUSICAL achieves a speed-up of up to 30.36x on a commodity machine with 32 cores with an efficiency of 94.88%. The experimental results also show that our new parallel MUSICAL outperforms the previous versions of MUSICAL in Matlab, Java, and Python by 30.43x, 2.63x, and 1.69x, respectively, on commodity machines.

Lambert probleminin modifiye Chebyshev-Picard yineleme yöntemini kullanarak paralel çözümü

“…Although it seems a high speed-up of 34 is achieved, we note that MATLAB programs are known to be much slower than similar C programs. For instance, one recent study [20] reports that their MATLAB program is approximately 30 times slower than the corresponding C++ program.…”

Automatic algorithm for determining bone and soft-tissue factors in dual-energy subtraction chest radiography

“…Those techniques are Intel Threading Building Blocks (TBB) and C Basic Linear Algebra Subprograms (CBLAS), which belong to the Intel Math Kernel Library (MKL) for large matrix computation, and programming optimization in C++. Those techniques have been applied successfully with an efficiency of 94.88% for reducing runtime [32]. By those techniques, we expect the runtime will be significantly reduced when running all processes, including determining the bone and soft-tissue factors, DES, the registration, and the enhancement of soft-tissue image contrast.…”