| Autor: |
Pelit, Aykut, Özalp, Pelin, Taştekin, Bora, İbrikçi, Turgay, Öcal, Işıl |
| Přispěvatelé: |
Çukurova Üniversitesi, Tıp Fakültesi, Temel Tıp Bilimleri Bölümü, Pelit, Aykut, Özalp, Pelin, Taştekin, Bora, İbrikçi, Turgay, Öcal, Işıl |
| Jazyk: |
angličtina |
| Rok vydání: |
2020 |
| Předmět: |
|
| Popis: |
The aim of this study is to forecast with Artificial Neural Network method the isometric contractile performance of skeletal muscles using weight and biochemical parameters of the rats applied magnetic field with diabetes mellitus (DM+MF) and as a control group only induced-DM instead of sacrificing animals. The multilayer back-propagation algorithm in neural network toolbox compatible with Matlab 2013Rb was used in this study. %70 (N=28) and %30 (N=12) of the 40 rats were used respectively for training and testing. The Levenberg–Maquardth algorithm also known as trainLM learning function was used for training. As inputs to the artificial neural network was used weight, glucose, high-density lipoprotein (HDL), low-density protein (LDL) and triglyceride values of the rats. As outputs to the artificial neural network was used the isometric contractile performance of skeletal muscles. The relationship between inputs and outputs values was taught to the artificial neural network. Then this network has been tested with data that was not previously defined. According to the training and testing data, the relationship between actual and predicted values of the neural network was calculated such as the mean absolute error (MAE), mean square error (MSE) and correlation coefficient (R). Mean Absolute Error was measured 0.99 and 1.07, respectively for training and testing. Mean Squared Error was measured 1.47 and 1.91, respectively for training and testing. Correlation coefficient was measured 0.92 and 0.98, respectively for training and testing. It was not found significant difference in terms of statistically between the actual values and estimated values of isometric contraction performance in rat skeletal muscles (p |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
