Genetic algorithms (GAs) are stochastic methods that are widely used in search and optimization. The breeding process is the main driving mechanism for GAs that leads the way to find the global optimum. And the initial phase of the breeding process starts with parent selection. The selection utilized in a GA is effective on the convergence speed of the algorithm. A GA can use different selection mechanisms for choosing parents from the population and in many applications the process generally depends on the fitness values of the individuals. Artificial neural networks (ANNs) are used to decide the appropriate parents by the new hybrid algorithm proposed in this study. And the use of neural networks aims to produce better offspring during the GA search. The neural network utilized in this algorithm tries to learn the structural patterns and correlations that enable two parents to produce high-fit offspring. In the breeding process, the first parent is selected based on the fitness value as usual. Then it is the neural network that decides the appropriate mate for the first parent chosen. Hence, the selection mechanism is not solely dependent on the fitness values in this study. The algorithm is tested with seven benchmark functions. It is observed from results of these tests that the new selection method leads genetic algorithm to converge faster.

No abstract

Background and Purpose: Despite continuing efforts in the multimodal assessment of the motor system after stroke, conclusive findings on the complementarity of functional and structural metrics of the ipsilesional corticospinal tract integrity and the role of the contralesional hemisphere are still lacking. This research aimed to find the best combination of motor system metrics, allowing the classification of patients into 3 predefined groups of upper limb motor recovery. Methods: We enrolled 35 chronic ischemic stroke patients (mean 47 [26–66] years old, 29 [6–58] months poststroke) with a single supratentorial lesion and unilateral upper extremity weakness. Patients were divided into 3 groups, depending on upper limb motor recovery: good, moderate, and bad. Nonparametric statistical tests and regression analysis were used to investigate the relationships among microstructural (fractional anisotropy (FA) ratio of the corticospinal tracts at the internal capsule (IC) level (classic method) and along the length of the tracts (Fréchet distance), and of the corpus callosum) and functional (motor evoked potentials [MEPs] for 2 hand muscles) motor system metrics. Stratification rules were also tested using a decision tree classifier. Results: IC FA ratio in the IC and MEP absence were both equally discriminative of the bad motor outcome (96% accuracy). For the 3 recovery groups’ classification, the best parameter combination was IC FA ratio and the Fréchet distance between the contralesional and ipsilesional corticospinal tract FA profiles (91% accuracy). No other metrics had any additional value for patients’ classification. MEP presence differed for 2 investigated muscles. Conclusions: This study demonstrates that better separation between 3 motor recovery groups may be achieved when considering the similarity between corticospinal tract FA profiles along its length in addition to region of interest-based assessment and lesion load calculation. Additionally, IC FA ratio and MEP absence are equally important markers for poor recovery, while for MEP probing it may be important to investigate more than one hand muscle.

No abstract

No abstract