TY  - THES
AU  - Millen, Nada
PY  - 2019
UR  - http://nardus.mpn.gov.rs/handle/123456789/12187
UR  - http://eteze.bg.ac.rs/application/showtheses?thesesId=7337
UR  - https://fedorabg.bg.ac.rs/fedora/get/o:21573/bdef:Content/download
UR  - http://vbs.rs/scripts/cobiss?command=DISPLAY&base=70036&RID=2048436834
UR  - https://farfar.pharmacy.bg.ac.rs/handle/123456789/3692
AB  - Vlažna granulacija je složen proizvodni proces izrade tableta. Ovaj proces je podložan uticaju mnogobrojnih varijabli, što predstavlja izazov prilikom transfera proizvodne tehnologije. Predmet istraživanja doktorske disertacije je olakšanje pocesa transfera tehnologije vlažne granulacije korišćenjem pristupa principa dizajna kvaliteta (engl. Quality by design – QbD) i različitih algoritama mašinskog učenja.Prvi deo ovog rada istražuje značaj i obim uticaja formulacionih i procesnih promenljivih na kritične atribute kvaliteta (engl. Critical quality attributes – CQA) tableta. Istraživan je uticaj nerastvornog punioca trikalcijum-fosfata (TKF) na vreme raspadanja i čvrstinu tableta u procesu vlažne granulacije. Ovaj ekscipijens nije prethodno izučavan u ovom kontekstu zbog toga što se u farmaceutskom razvoju najčešće zamenjuje dikalcijum-fosfatom. Uticaj koncentracije trikalcijum-fosfata i superdezintegratora natrijum-skrobglikolata (NSG) i uticaj procesnih parametara (količina vode u rastvoru za granulaciju, veličina otvora sita i sadržaj vlage) su procenjeni istovremeno u cilju razvoja matematičkih modela za predviđanje kritičnih atributa kvaliteta tableta (vreme raspadanja i čvrstina) i uspešnog vođenja procesa transfera tehnologije vlažne granulacije. Povećana koncentracija natrijum-skrobglikolata je skratila vreme raspadanja ali je imala negativan uticaj na čvrstinu tableta. Procesni parametar sa najviše uticaja na vreme raspadanja je bio sadržaj vlage dok je veličina otvora na situ (koje je korišćeno prilikom mlevenja suvog granulata) bio najznačajniji procesni parametar za čvrstinu tableta. Pri upotrebi sita sa manjom veličinom otvora u procesu mlevenja dobijeni su optimalni rezultati za analizirane atribute kvaliteta tableta kad je količina vode u rastvoru za granulaciju bila na višem nivou. Kada je za mlevenje granula korišćeno sito sa većom veličinom otvora, najbolji rezultati za atribute kvaliteta tableta su dobijeni kad je količina vode u rastvoru za granulaciju bila na nižem nivou. Razvijeni su matematički modeli koji opisuju uticaj ispitivanih faktora na vreme raspadanja i čvrstinu tableta. Pouzdanost matematičkih modela je procenjivana tokom procesa transfera sa laboratorijskog na komercijalni nivo proizvodnje...
AB  - Wet granulation is a complex manufacturing process. It is influenced by the multiple variables which presents challenges to the scale-up process. The objective of this PhD dissertation research is to use a quality by design (QbD) approach i multiple machine learning algorithms in facilitating wet granulation process scale-up.The first part of this reseach investigates the significance i extent of influence of both formulation i process variables on the critical quality attributes (CQAs) of tablets. The influence of insoluble diluent - tribasic calcium phosphate on disintegration time i tablet hardness, in the wet granulation process, was examined. This excipient was not previously studied in this context as it is frequently replaced with dicalcium phosphate in the pharmaceutical development practice. The influence of tricalcium phosphate i the superdisintegrant sodium starch glycolate, as well as the influence of process parameters (water concentration in granulation solution, screen hole size i moisture content) were used together to develop mathematical models that could be used in prediction of tablet CQAs (disintegration time i hardness) i facilitate wet granulation scale up process. Higher concentrations of sodium starch glycolate shortened disintegration time but negatively influenced tablet hardness. The most significant factor influencing disintegration time was moisture content while a screen hole size (used for dry granulate milling) was the most significant factor influencing tablet hardness. The use of a smaller screen hole size in milling process produced the optimal results for tablet quality attributes when water amount in granulation solution was high. The best results for tablet quality attributes are achieved when a larger hole screen size, used in milling process, was applied in conjuction with the low water amount, used in granulation process. Developed mathematical models describe the influence of analysed parameters on disintegration time i tablet hardness. Those equations were evaluated during scale-up process from laboratory to commercial scale. Experimental i predicted mean values were found to be statistically similar, which proves the effectiveness of the models in scale-up projects.In the second part of this research the aim was to further investigate the extent of influence of critical variables through application of multipe machine learning algorithms...
PB  - Универзитет у Београду, Фармацеутски факултет
T2  - Универзитет у Београду
T1  - Razvoj multidimenzionalnog modela za predviđanje i kontrolu ciljnog profila kvaliteta tableta izrađenih postupkom vlažne granulacije
UR  - https://hdl.handle.net/21.15107/rcub_nardus_12187
ER  - 

@phdthesis{
author = "Millen, Nada",
year = "2019",
abstract = "Vlažna granulacija je složen proizvodni proces izrade tableta. Ovaj proces je podložan uticaju mnogobrojnih varijabli, što predstavlja izazov prilikom transfera proizvodne tehnologije. Predmet istraživanja doktorske disertacije je olakšanje pocesa transfera tehnologije vlažne granulacije korišćenjem pristupa principa dizajna kvaliteta (engl. Quality by design – QbD) i različitih algoritama mašinskog učenja.Prvi deo ovog rada istražuje značaj i obim uticaja formulacionih i procesnih promenljivih na kritične atribute kvaliteta (engl. Critical quality attributes – CQA) tableta. Istraživan je uticaj nerastvornog punioca trikalcijum-fosfata (TKF) na vreme raspadanja i čvrstinu tableta u procesu vlažne granulacije. Ovaj ekscipijens nije prethodno izučavan u ovom kontekstu zbog toga što se u farmaceutskom razvoju najčešće zamenjuje dikalcijum-fosfatom. Uticaj koncentracije trikalcijum-fosfata i superdezintegratora natrijum-skrobglikolata (NSG) i uticaj procesnih parametara (količina vode u rastvoru za granulaciju, veličina otvora sita i sadržaj vlage) su procenjeni istovremeno u cilju razvoja matematičkih modela za predviđanje kritičnih atributa kvaliteta tableta (vreme raspadanja i čvrstina) i uspešnog vođenja procesa transfera tehnologije vlažne granulacije. Povećana koncentracija natrijum-skrobglikolata je skratila vreme raspadanja ali je imala negativan uticaj na čvrstinu tableta. Procesni parametar sa najviše uticaja na vreme raspadanja je bio sadržaj vlage dok je veličina otvora na situ (koje je korišćeno prilikom mlevenja suvog granulata) bio najznačajniji procesni parametar za čvrstinu tableta. Pri upotrebi sita sa manjom veličinom otvora u procesu mlevenja dobijeni su optimalni rezultati za analizirane atribute kvaliteta tableta kad je količina vode u rastvoru za granulaciju bila na višem nivou. Kada je za mlevenje granula korišćeno sito sa većom veličinom otvora, najbolji rezultati za atribute kvaliteta tableta su dobijeni kad je količina vode u rastvoru za granulaciju bila na nižem nivou. Razvijeni su matematički modeli koji opisuju uticaj ispitivanih faktora na vreme raspadanja i čvrstinu tableta. Pouzdanost matematičkih modela je procenjivana tokom procesa transfera sa laboratorijskog na komercijalni nivo proizvodnje..., Wet granulation is a complex manufacturing process. It is influenced by the multiple variables which presents challenges to the scale-up process. The objective of this PhD dissertation research is to use a quality by design (QbD) approach i multiple machine learning algorithms in facilitating wet granulation process scale-up.The first part of this reseach investigates the significance i extent of influence of both formulation i process variables on the critical quality attributes (CQAs) of tablets. The influence of insoluble diluent - tribasic calcium phosphate on disintegration time i tablet hardness, in the wet granulation process, was examined. This excipient was not previously studied in this context as it is frequently replaced with dicalcium phosphate in the pharmaceutical development practice. The influence of tricalcium phosphate i the superdisintegrant sodium starch glycolate, as well as the influence of process parameters (water concentration in granulation solution, screen hole size i moisture content) were used together to develop mathematical models that could be used in prediction of tablet CQAs (disintegration time i hardness) i facilitate wet granulation scale up process. Higher concentrations of sodium starch glycolate shortened disintegration time but negatively influenced tablet hardness. The most significant factor influencing disintegration time was moisture content while a screen hole size (used for dry granulate milling) was the most significant factor influencing tablet hardness. The use of a smaller screen hole size in milling process produced the optimal results for tablet quality attributes when water amount in granulation solution was high. The best results for tablet quality attributes are achieved when a larger hole screen size, used in milling process, was applied in conjuction with the low water amount, used in granulation process. Developed mathematical models describe the influence of analysed parameters on disintegration time i tablet hardness. Those equations were evaluated during scale-up process from laboratory to commercial scale. Experimental i predicted mean values were found to be statistically similar, which proves the effectiveness of the models in scale-up projects.In the second part of this research the aim was to further investigate the extent of influence of critical variables through application of multipe machine learning algorithms...",
publisher = "Универзитет у Београду, Фармацеутски факултет",
journal = "Универзитет у Београду",
title = "Razvoj multidimenzionalnog modela za predviđanje i kontrolu ciljnog profila kvaliteta tableta izrađenih postupkom vlažne granulacije",
url = "https://hdl.handle.net/21.15107/rcub_nardus_12187"
}

Millen, N.. (2019). Razvoj multidimenzionalnog modela za predviđanje i kontrolu ciljnog profila kvaliteta tableta izrađenih postupkom vlažne granulacije. in Универзитет у Београду
Универзитет у Београду, Фармацеутски факултет..
https://hdl.handle.net/21.15107/rcub_nardus_12187

Millen N. Razvoj multidimenzionalnog modela za predviđanje i kontrolu ciljnog profila kvaliteta tableta izrađenih postupkom vlažne granulacije. in Универзитет у Београду. 2019;.
https://hdl.handle.net/21.15107/rcub_nardus_12187 .

Millen, Nada, "Razvoj multidimenzionalnog modela za predviđanje i kontrolu ciljnog profila kvaliteta tableta izrađenih postupkom vlažne granulacije" in Универзитет у Београду (2019),
https://hdl.handle.net/21.15107/rcub_nardus_12187 .

TY  - JOUR
AU  - Millen, Nada
AU  - Kovačević, A
AU  - Đuriš, Jelena
AU  - Ibrić, Svetlana
PY  - 2019
UR  - https://farfar.pharmacy.bg.ac.rs/handle/123456789/3292
AB  - Purpose: Optimal particle size distribution (PSD) is an important factor in wet granulation in order to achieve appropriate powder flow, compactibility, and content uniformity. Parameters like D50 and surface area (SA) are used to define PSD but both are only able to compare separate fractions of a granulate. In this work, we made an attempt to characterize PSD of a final dry granulate blend and suggest novel parameters (determination coefficient R2 and trend line slope of a PSD model) to quantitatively describe PSD. Method: The significance of these parameters was tested using machine learning. Laboratory-scale samples were used for training and commercial-scale samples for testing a model. Several machine learning techniques were used to further examine the importance of these input variables using a large data set from wet granulation scale-up study. Results: The Gradient Boosted Regression Trees (GBRT) algorithm had the lowest root mean square error (RMSE) values for the several responses studied (tablet tensile strength, tablet diameter and thickness, compaction work, decompaction work, and net work). The GBRT model for tablet tensile strength had an R2 model value of 0.87 and was not overfitted. The importance of input variables R2 and a was proven by the stepwise regression model’s p value (0.0003) and GBRT importance score (0.37 and 0.44, respectively). The GBRT model was the most successful in predicting decompaction work (R2 model = 0.97) with the least regularization effect. Conclusion: The proposed parameters can be used in PSD characterization and applied in critical quality attributes (CQA) prediction and wet granulation scale-up.
PB  - Springer New York LLC
T2  - Journal of Pharmaceutical Innovation
T1  - Machine Learning Modeling of Wet Granulation Scale-up Using Particle Size Distribution Characterization Parameters
DO  - 10.1007/s12247-019-09398-0
ER  - 

@article{
author = "Millen, Nada and Kovačević, A and Đuriš, Jelena and Ibrić, Svetlana",
year = "2019",
abstract = "Purpose: Optimal particle size distribution (PSD) is an important factor in wet granulation in order to achieve appropriate powder flow, compactibility, and content uniformity. Parameters like D50 and surface area (SA) are used to define PSD but both are only able to compare separate fractions of a granulate. In this work, we made an attempt to characterize PSD of a final dry granulate blend and suggest novel parameters (determination coefficient R2 and trend line slope of a PSD model) to quantitatively describe PSD. Method: The significance of these parameters was tested using machine learning. Laboratory-scale samples were used for training and commercial-scale samples for testing a model. Several machine learning techniques were used to further examine the importance of these input variables using a large data set from wet granulation scale-up study. Results: The Gradient Boosted Regression Trees (GBRT) algorithm had the lowest root mean square error (RMSE) values for the several responses studied (tablet tensile strength, tablet diameter and thickness, compaction work, decompaction work, and net work). The GBRT model for tablet tensile strength had an R2 model value of 0.87 and was not overfitted. The importance of input variables R2 and a was proven by the stepwise regression model’s p value (0.0003) and GBRT importance score (0.37 and 0.44, respectively). The GBRT model was the most successful in predicting decompaction work (R2 model = 0.97) with the least regularization effect. Conclusion: The proposed parameters can be used in PSD characterization and applied in critical quality attributes (CQA) prediction and wet granulation scale-up.",
publisher = "Springer New York LLC",
journal = "Journal of Pharmaceutical Innovation",
title = "Machine Learning Modeling of Wet Granulation Scale-up Using Particle Size Distribution Characterization Parameters",
doi = "10.1007/s12247-019-09398-0"
}

Millen, N., Kovačević, A., Đuriš, J.,& Ibrić, S.. (2019). Machine Learning Modeling of Wet Granulation Scale-up Using Particle Size Distribution Characterization Parameters. in Journal of Pharmaceutical Innovation
Springer New York LLC..
https://doi.org/10.1007/s12247-019-09398-0

Millen N, Kovačević A, Đuriš J, Ibrić S. Machine Learning Modeling of Wet Granulation Scale-up Using Particle Size Distribution Characterization Parameters. in Journal of Pharmaceutical Innovation. 2019;.
doi:10.1007/s12247-019-09398-0 .

Millen, Nada, Kovačević, A, Đuriš, Jelena, Ibrić, Svetlana, "Machine Learning Modeling of Wet Granulation Scale-up Using Particle Size Distribution Characterization Parameters" in Journal of Pharmaceutical Innovation (2019),
https://doi.org/10.1007/s12247-019-09398-0 . .

TY  - JOUR
AU  - Millen, Nada
AU  - Kovacević, Aleksandar
AU  - Khera, Lalit
AU  - Đuriš, Jelena
AU  - Ibrić, Svetlana
PY  - 2019
UR  - https://farfar.pharmacy.bg.ac.rs/handle/123456789/3245
AB  - The purpose of this extensive study is to use a quality by design (QbD) approach and multiple machine learning algorithms in facilitating wet granulation process scale-up. This study investigated the extent of influence of both formulation and process variables. Furthermore, measured responses covered compressibility, compactibility and manufacturability of a powder blend. Finally, the models developed on laboratory scale samples were tested on pilot and commercial scale runs. Tablet detachment and ejection work were calculated from force-displacement measurements. Significant numerical and categorical input variables were identified by using a stepwise regression model and their importance evaluated by using a boosted trees model. Pilot scale runs resulted in the highest tablet tensile strength and compaction work as well as the highest detachment and ejection work. Critical quality attributes (CQAs) that were the most successfully predicted were the compaction, decompaction, and net work, as well as the tablet height. The most important input variable influencing all CQAs was the compaction force. Application of the boosted regression trees model resulted in the lowest Root Mean Square Error (RMSE) values for all of the responses. This work demonstrates reliability of predictions of developed models that can be successfully used as a part of a QbD approach for wet granulation scale-up.
PB  - Savez hemijskih inženjera, Beograd
T2  - Hemijska industrija
T1  - Machine learning modelling of wet granulation scale-up using compressibility, compactibility and manufacturability parameters
VL  - 73
IS  - 3
SP  - 155
EP  - 168
DO  - 10.2298/HEMIND190412017M
ER  - 

@article{
author = "Millen, Nada and Kovacević, Aleksandar and Khera, Lalit and Đuriš, Jelena and Ibrić, Svetlana",
year = "2019",
abstract = "The purpose of this extensive study is to use a quality by design (QbD) approach and multiple machine learning algorithms in facilitating wet granulation process scale-up. This study investigated the extent of influence of both formulation and process variables. Furthermore, measured responses covered compressibility, compactibility and manufacturability of a powder blend. Finally, the models developed on laboratory scale samples were tested on pilot and commercial scale runs. Tablet detachment and ejection work were calculated from force-displacement measurements. Significant numerical and categorical input variables were identified by using a stepwise regression model and their importance evaluated by using a boosted trees model. Pilot scale runs resulted in the highest tablet tensile strength and compaction work as well as the highest detachment and ejection work. Critical quality attributes (CQAs) that were the most successfully predicted were the compaction, decompaction, and net work, as well as the tablet height. The most important input variable influencing all CQAs was the compaction force. Application of the boosted regression trees model resulted in the lowest Root Mean Square Error (RMSE) values for all of the responses. This work demonstrates reliability of predictions of developed models that can be successfully used as a part of a QbD approach for wet granulation scale-up.",
publisher = "Savez hemijskih inženjera, Beograd",
journal = "Hemijska industrija",
title = "Machine learning modelling of wet granulation scale-up using compressibility, compactibility and manufacturability parameters",
volume = "73",
number = "3",
pages = "155-168",
doi = "10.2298/HEMIND190412017M"
}

Millen, N., Kovacević, A., Khera, L., Đuriš, J.,& Ibrić, S.. (2019). Machine learning modelling of wet granulation scale-up using compressibility, compactibility and manufacturability parameters. in Hemijska industrija
Savez hemijskih inženjera, Beograd., 73(3), 155-168.
https://doi.org/10.2298/HEMIND190412017M

Millen N, Kovacević A, Khera L, Đuriš J, Ibrić S. Machine learning modelling of wet granulation scale-up using compressibility, compactibility and manufacturability parameters. in Hemijska industrija. 2019;73(3):155-168.
doi:10.2298/HEMIND190412017M .

Millen, Nada, Kovacević, Aleksandar, Khera, Lalit, Đuriš, Jelena, Ibrić, Svetlana, "Machine learning modelling of wet granulation scale-up using compressibility, compactibility and manufacturability parameters" in Hemijska industrija, 73, no. 3 (2019):155-168,
https://doi.org/10.2298/HEMIND190412017M . .