AUTHOR=Li Feng , Xu Wensheng , Feng Yujin , Wang Wengang , Tian Hui , He Suhuan , Li Liang , Xiang Bai , Wang Yueheng TITLE=Preparation of ultrasound contrast agents: The exploration of the structure-echogenicity relationship of contrast agents based on neural network model JOURNAL=Frontiers in Oncology VOLUME=Volume 12 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.964314 DOI=10.3389/fonc.2022.964314 ISSN=2234-943X ABSTRACT=22 There is a need to standardize the process of micro/nanobubble preparation to 23 bring it closer to clinical translation. We explored a neural network-based 24 model to predict the structure-echogenicity relationship for the preparation 25 and fabrication of ultrasound-enhanced contrast agents. Seven formulations 26 were screened, and 109 measurements were obtained. An artificial neural 27 network-multilayer perceptron (ANN-MLP) model was used. The original data 28 were divided into training and testing groups, which included 73 and 36 groups 29 of data, respectively. The hidden layer was selected from three hidden layers 30 and included bias. The classification graph showed that the predicted values of 31 the training and testing groups were 76.7% and 66.7%, respectively. According 32 to the receiver operating characteristic curve, the accuracy of different 33 imaging effects could achieve a prediction rate of 88.1–96.5%. The percentage 34 graph showed that the data were gradually converging. The predictive analysis 35 curves of different ultrasound effects gradually approached a plane. 36 Normalized importance predicted contributions for the Pk1, poly-dispersity 37 index (PDI), and intensity account were 100%, 98.5%, and 98.7%, respectively. 38 Application of the ANN-MLP model is feasible and effective in exploring the 39 synthesis process of ultrasound contrast agents. 1,2-Distearoyl-sn-glycero-3- 40 phosphoethanolamine-N-(methoxy[polyethylene glycol]-2000) (DSPE-PEG-2000) 41 correlated highly with the success rate of contrast agent synthesis.