Data-driven insights into the characteristics of liquisolid systems based on the machine learning algorithms DOI Creative Commons
Ivana Vasiljević, Erna Turković, Jelena Parojčić

и другие.

European Journal of Pharmaceutical Sciences, Год журнала: 2024, Номер 203, С. 106927 - 106927

Опубликована: Окт. 6, 2024

Liquisolid systems (LS) represent a formulation approach where liquid drug or its dispersion is transformed into powder with good flowability and compactibility, leading to enhanced dissolution bioavailability. Many research groups have focused on the preparation investigation of LS, higher need for comprehensive evaluation factors impacting LS characteristics. The aim this work was investigate applicability machine learning algorithms in evaluation, using data mined from published literature, provide an insight critical governing liquisolid system performance. dataset prepared publication search engines relevant keywords, total 425 formulations included database. database methods, parameters, Subsequently, properties system, i.e. flowability, compact hardness, dissolution, were analyzed algorithms, including Gradient Boosting, Adaptive Boosting Random Forest. In addition conventional methods excipients, novel technologies (fluid bed preparation, extrusion/spheronization) materials (Neusilin®, Fujicalin®, Syloid®) systems. analysis revealed that factors, such as carrier coating agent type content, phase load, model well method, significantly influenced models developed exhibited high prediction accuracy when applied test (higher than 80 %). This indicates may attributes affecting performance be used valuable tool development optimization these samples.

Язык: Английский

Data-driven insights into the characteristics of liquisolid systems based on the machine learning algorithms DOI Creative Commons
Ivana Vasiljević, Erna Turković, Jelena Parojčić

и другие.

European Journal of Pharmaceutical Sciences, Год журнала: 2024, Номер 203, С. 106927 - 106927

Опубликована: Окт. 6, 2024

Liquisolid systems (LS) represent a formulation approach where liquid drug or its dispersion is transformed into powder with good flowability and compactibility, leading to enhanced dissolution bioavailability. Many research groups have focused on the preparation investigation of LS, higher need for comprehensive evaluation factors impacting LS characteristics. The aim this work was investigate applicability machine learning algorithms in evaluation, using data mined from published literature, provide an insight critical governing liquisolid system performance. dataset prepared publication search engines relevant keywords, total 425 formulations included database. database methods, parameters, Subsequently, properties system, i.e. flowability, compact hardness, dissolution, were analyzed algorithms, including Gradient Boosting, Adaptive Boosting Random Forest. In addition conventional methods excipients, novel technologies (fluid bed preparation, extrusion/spheronization) materials (Neusilin®, Fujicalin®, Syloid®) systems. analysis revealed that factors, such as carrier coating agent type content, phase load, model well method, significantly influenced models developed exhibited high prediction accuracy when applied test (higher than 80 %). This indicates may attributes affecting performance be used valuable tool development optimization these samples.

Язык: Английский

Процитировано

0