Fuzzy dynamic parameter adaptation in the bird swarm algorithm for neural network optimization.

Fuzzy dynamic parameter adaptation has proven to be of great help when it is implemented in bio-inspired algorithms for optimization in different application areas, such as control, mathematical functions, classification, among others. One of the main contributions of this work is the proposed improvement of the Bird Swarm algorithm using a Fuzzy System approach, and we called this improvement the Fuzzy Bird Swarm Algorithm. Furthermore, we use a set of complex Benchmark Functions of the Congress on Evolutionary Computation Competition 2017 to compare the results between the original algorithm and the proposed improvement of the algorithm. The fuzzy system is utilized for the dynamic parameter adaptation of the C1 and C2 parameters of the Bird Swarm Algorithm. As a result, the Fuzzy Bird Swarm Algorithm has enhanced exploration and exploitation abilities that help in achieving better results than the Bird Swarm Algorithm. We additionally test the algorithm's performance in a real problem in the medical area, using the optimization of a neural network to obtain the risk of developing hypertension. This neural network uses information, such as age, gender, body mass index, systolic pressure, diastolic pressure, if the patient smokes and if the patient has parents with hypertension. Hypertension is one of the leading causes of heart problems, which in turn are also one of the top causes of death. Moreover, these days it causes more complications and deaths in people infected with COVID-19, the virus of the ongoing pandemic. Based on the results obtained through the 30 experiments carried out in three different study cases, and the results obtained from the statistical tests, it can be concluded that the proposed method provides better performance when compared with the original method.

Blood Pressure Classification Using the Method of the Modular Neural Networks.

In this paper, we present a new model based on modular neural networks (MNN) to classify a patient's blood pressure level (systolic and diastolic pressure and pulse). Tests are performed with the Levenberg-Marquardt (trainlm) and scaled conjugate gradient backpropagation (traincsg) training methods. The modular neural network architecture is formed by three modules. In the first module we consider the diastolic pressure data; in the second module we use details of the systolic pressure; in the third module, pulse data is used and the response integration is performed with the average method. The goal is to design the best MNN architecture for achieving an accurate classification. The results of the model show that MNN presents an excellent classification for blood pressure. The contribution of this work is related to helping the cardiologist in providing a good diagnosis and patient treatment and allows the analysis of the behavior of blood pressure in relation to the corresponding diagnosis, in order to prevent heart disease.