| BackgroundThe antibacterial method induced by microbubble-enhanced sonoporation has shown its great potential in facilitating drug delivery into thallus.The enhanced drug delivery induced by microbubble-enhanced sonoporation is a complex event which can be affected by various physical parameters.How to determine the correlation between experimental parameters and the drug delivery efficiency to give the instruction on reasonably choosing the parameters and achieve the control of drug delivery efficiency is impeding further investigations for this complex biophysical process.Tuberculosis(TB)is the most deadly infectious disease in the world.Its pathogen mycobacterium tuberculosis(MTB)has a thick cell wall and poor permeability,and it mainly parasitizes in macrophages(Mф),which results in drug resistance and double barriers to drug penetration into MTB.Through artificial neural network modeling for the enhanced drug delivery effect induced by microbubble-enhanced sonoporation,we provide a quantitative framework for further understanding of this complex biophysical process.Our model offers a predictive tool for drug delivery efficiency under different experimental parameters.It is expected to provide a theoretical solution for the parameter optimization of the antibacterial method induced by microbubble-enhanced sonoporation.ObjectiveThrough artificial neural network modeling for the enhanced drug delivery into bacteria induced by microbubble-enhanced sonoporation,we realize the prediction of drug delivery efficiency under different experimental parameters,and then achieve the control of drug delivery efficiency through reasonable parameter selection.Methods1.Artificial neural network identification for the damage effect of ultrasonic cavitation on macrophage:In our previous research,we studied the damage effects of low-frequency and low-intensity ultrasound(LFLIU)on macrophage.The ultrasound frequency is 42 kHz,and the ultrasonic intensity is in the range of 0.13 to 0.34 W/cm~2.Based on the experimental data,we aim to use modified artificial neural network to establish the quantitative relationship between acoustic parameters and cavitation effect,which will lay a theoretical foundation for the model identification of the enhanced drug delivery effect induced by microbubble-enhanced sonoporation in the next step of our research.2.Artificial neural network identification for the enhanced drug delivery into M.smegmatis induced by microbubble-enhanced sonoporation:The enhanced drug delivery into M.smegmatis induced by microbubble-enhanced sonoporation is a complex event which can be affected by various physical parameters.In this paper,we explored a number of key parameters affecting the drug delivery efficiency induced by microbubble-enhanced sonoporation.The experiment selected an ultrasonic transmitter with a fixed frequency of 42 kHz,transducer diameter of 1 cm,output sound intensity range of 0.15 to 0.6 W/cm~2.The bacterial survival rate was used to indicate the drug delivery efficiency induced by microbubble-enhanced sonoporation.Low bacterial survival rate indicates efficient drug delivery results.Based on the biological experiment data,a modified ANN model is constructed to identify the quantitative relationship between experimental parameters and drug delivery efficiency.Due to the nonlinearity of drug delivery system,poor selection of training samples will cause model mismatch.Also,it is difficult to determine the number of training samples through a random selection method,and there is a strong uncertainty.Therefore,the multiple model idea was introduced into the selection of training samples to modify the traditional back-propagation neural network model to avoid model mismatch caused by poor training sample selection,which provides a theoretical basis for the selection of training samples and achieve the improvement of traditional neural network model.By analyzing the experimental samples,a mapping relationship expression can be deduced to determine the input and output variables of artificial neural network models.Experimental samples were divided into training and test samples.We trained models based on back-propagation neural network to establish their quantitative relationship.Results1.Artificial neural network identification for the damage effect of ultrasonic cavitation on macrophage:Based on the combination of the multiple model idea and artificial neural network,the identification model of cavitation effect on macrophage was constructed.It is elucidated that an appropriately trained network can act as a good alternative for costly and time-consuming experiments(Error Index EI=0.0137;Prediction Accuracy PA=100%).2.Artificial neural network identification for the enhanced drug delivery into M.smegmatis induced by microbubble-enhanced sonoporation:In the process of the synergistic bactericidal process induced by microbubble-enhanced sonoporation,with the introduction of microbubbles and the increase of ultrasonic intensity,ultrasonic irradiation time and levofloxacin concentration,the drug delivery efficiency become higher.Based on the experimental data,the modified artificial neural network model build in this paper has high identification accuracy.This implies that the new method for training sample selection proposed in this paper can provide a theoretical basis for the selection of training samples and realize the improvement of traditional neural network model to avoid model mismatch caused by unreasonable training sample selection.(traditional ANN model:Error Index EI=0.5018,Prediction Accuracy PA=11.11%;modified ANN model:Error Index EI=0.0298,Prediction Accuracy PA=90.54%).Conclusions1.The new method for training sample selection proposed in this paper can provide a theoretical basis for the selection of training samples and realize the improvement of traditional neural network model to avoid model mismatch caused by unreasonable training sample selection.2.The findings of this study indicate that this approach can realize the prediction of drug delivery efficiency induced by microbubble-enhanced sonoporation under different experimental parameters,and then achieve the control of drug delivery efficiency through reasonable parameter selection,so as to effectively kill mycobacterium smegmatis. |
PDF Full Text Request