Investigation of Optimal Heuristical Parameters for Mixed ACO-k-means Segmentation Algorithm for MRI Images

Abstract

-The parameters of the modified mixed Ant Colony Optimization (ACO) -k-means image segmentation algorithm are considered. There have been investigated such parameters as n -the number of ants; heuristic coefficients of ACO algorithm and their dependence on the image scale and number of iterations before and after parameters correction. The proposed algorithm and sub-system for the study of coefficients, as part of the medical image segmentation system, have been implemented. Operation of the algorithm with and without the use of optimal parameters was applied. Optimal parameters were studied for 6 groups of MRI images: brain, heart, lungs, liver, bone structures, and others. The results are displayed in the final table. Images from Ossirix image dataset and real patients' images were used for testing.

Exponential Particle Swarm Optimization Algorithm for Complexly Structured Images Segmentation

“…The PSO approach employs a collection of particles, each of which has a unique local solution [3,4]. According to its own habits and those of its neighbors, the particle's behavior varies every time it enters the search zone.…”

Recommendation Assistant System for Social Networks and Search Services Based on Population Filtering Algorithm

“…The proposed hybrid filtration model is based on the application of the population algorithm as a machine learning method to solve the problem [9,10]. A variety of input data in the hybrid model is supported by a population of custom "characteristics" encoded in a population algorithm.…”

“…Note that this result was obtained under the assumption that the number of generations (which is equivalent to the number of calculations of the objective function) is the most important factor in estimating the computation time of the algorithm. This is probably true for most applications of population-based algorithms because the estimation of the objective function there is the most time-consuming part of the algorithm, in contrast to the complexity of executing the operators of population-based algorithms, the complexity of which is estimated as O(n) -O(n ln(n)) [9]. How are solutions have encoded?…”