Information Technology Reference
In-Depth Information
Optimization of the tablet formulation.
ANN inputs were formulation composition,
dwell time, and compression force; whereas
outputs were normalized ejection and
residual forces.
MLP
Bourquin
et al., 1998
MLP
Leonardi
et al., 2009
Optimization of benzidazole chitosan
microparticles. Inputs for the network training
were the polymer concentration, the NaOH
concentration, the stirring rate, and the spraying
rate (microparticles were prepared by the
coacervation method). Outputs for the
network training were microparticle size,
encapsulation effi ciency, yield, and
dissolution rates.
Analysis of diffusion coeffi cients from
mucoadhesive vaginal controlled drug delivery
system. Inputs for the network training were
sodium dodecyl sulfate dose, delivery device
weight, vaginal fl uid simulant pH, fl ow rate, and
speed of rotation. Release profi les (i.e. diffusion
coeffi cients) were selected as outputs for the
network training.
MLP
Lee et al.,
2008
Viscosity prediction of lipophilic semisolid
emulsion systems. Formulation composition was
used as input in the network training, whereas
parameters indicative of formulations viscosity
were outputs.
MLP
Gašperlin
et al., 2000
Prediction of microemulsion phase boundaries in
PEG-8 caprylic/capric glycerides based systems.
Input parameters for the network training were
the surfactant concentration in surfactant/
cosurfactant mixture and the oil concentration in
the mixture with tensides. Output parameter was
the water solubilization limit.
GRNN
Djekic
et al., 2008
Analysis of relationships between formulation
characteristics and pellet properties. Input
parameters used for the network training were
parameters describing pellet quantitative and
qualitative composition, as well as technology of
preparation, whereas output parameters were
pellet aspect ratio (AR) and mean dissolution
time (MDT).
MLP
Mendyk
et al., 2010
(
Continued
)
Search WWH ::
Custom Search