Design and evaluation of a parallel and optimized light–tissue interaction-based method for fast skin lesion assessment

Abstract

In recent years, image processing technics have attracted much attention as powerful tools in the assessment of skin lesions from multispectral images. The Kubelka-Munk Genetic Algorithm (KMGA) is a novel method which has been developed for this diagnostic purpose. It combines the Kubelka-Munk light-tissue interaction model with the Genetic Algorithm optimization process, and allows quantitative measure of cutaneous tissue by computing skin parameter maps such as melanin concentration, volume blood fraction, oxygen saturation or epidermis/dermis thickness. However, its efficiency is seriously reduced by the mass floating-point operations for each pixel of the multispectral image, and this prevents the algorithm from reaching industrial standards related to cost, power and speed for clinical applications. In this paper, our work focuses on the improvement of this theoretical achievement. Therefore, we repropose a new C-based Parallel and Optimized KMGA (PO-KMGA) technique designed and optimized using multiple ways: KM model optimized re-writing, operation massively parallelized using POSIX threads, memory use optimization and routine pipelining with Intel C?? Compiler, etc.Intensive experiments demonstrate that our introduced PO-KMGA framework spends less than 10 min to finish a job that the conventional KMGA spends around two days to do in the same hardware environment with a similar algorithm performance.

A predictive function optimization algorithm for multi-spectral skin lesion assessment

“…This results that the C symbolic expression of KM is quite inefficient to the compiler. Thus, we use a reduced KM model that is mathematically simplified in our previous work in order to reduce the unnecessary repetitive operations (see Section 3.1 of [9]). The computation cost of re-specified KM model is only 45.85% of its prototype.…”

High-level synthesis for FPGAs: code optimization strategies for real-time image processing

“…From the hardware point of view, the challenge of real-time image processing is to find the optimal platform satisfying the requirements of the image processing application among a large space of potential solutions. Its solution exploration therefore usually revolves around how to combine the software implementation with the hardware platform [1][2][3][4][5].…”

Embedded Implementation of VHR Satellite Image Segmentation

“…A series of highly parallelized image processing devices have been made available to engineers at a very affordable price. Such devices have been widely used in various signal processing and communication systems for their significant advantages in terms of running-cost, embeddability, power consumption or flexibility [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ]. For example, Zeng Yonghong [ 7 ] presented an efficient Intellectual Property (IP) core design methodology to implement a real-time image processing application, such as the Normalized Product correlation (NProd) image matching algorithm; Chiesi et al [ 2 ] proposed a new non-conventional technique based on Fuzzy Deconvolution for Scattering Center Detection (F-SCD) and its embedded implementation for real-time deployment in an automotive collision avoidance application.…”