“…Over the past decade, a multitude of computational studies have emerged focusing on predicting BBB permeability. Noteworthy contributions include works by Bujak et al (2015), 22 Zhang et al (2015), 23 Brito-S. Y. et al (2015), 24 Wang et al (2015), 25 Toropov et al (2017), 26 Radchenko et al (2020), 27 Shin et al (2021), 28 Wu et al (2021), 29 Kim et al (2021), 30 Radan et al (2022), 31 Tang et al (2022), 32 Shaker et al (2023), 33 35 These studies employ diverse algorithms, such as multiple linear regression (MLR), genetic algorithm-support vector machine (GA-SVM), 36 random forest (RF), 37 Monte Carlo (MC-SMILES), 38 artificial neural network (ANN), 39 Deep-B3, 32 support vector regression (SVR), 40 light gradient-boosting machine (LightGBM), 41 and linear discriminant analysis (LDA). 42 The datasets used in these investigations range from 18 to 7807 compounds, showcasing the breadth of approaches applied to understanding BBB permeability prediction.…”