Index – Computer-Aided Applications in Pharmaceutical Technology

Index

A

Advanced CAT (ACAT), 178, 180–3, 192–5, 202, 208
Activation function, 63, 94–6, 99, 102–3, 107–8
Analysis of variance (ANOVA), 21, 46–7, 50, 70
Area under curve (AUC), 188, 190, 197, 202, 204–5, 208, 214–15, 220, 223–4
Architecture (of ANN) See Topology
Artificial neural network (ANN), 12, 19–27, 63, 69, 75, 92–4, 96–101, 107–19, 122–3, 139, 144, 146, 149, 263

B

Back propagation, 22, 75, 97, 98, 105
Batch, 2, 6–7, 70–1, 73, 75, 78, 81, 83–5, 97, 118, 139, 143, 245
Bayesian, 12, 102
Binary logic, 119
Bioequivalence, 198
Biopharmaceutics
characterization, 178–9, 183, 210, 223, 226
classification system (BCS), 180, 185, 188–9, 192, 210, 221–4
Biowaiver, 179, 195, 210, 221–5
Box-Behnken design (BBD), 37–40, 53–4

C

C4.5 algorithm, 125–7, 130, 136
C5.0 algorithm, 125–6, 132
Capsules, 73, 192–5, 201–3, 208, 217–18
Causal factors, 21, 156–7
Central composite design (CCD), 21, 37–8, 40
Chemometric
model, 9, 59, 262
techniques (methods), 46, 57–8, 72, 77, 262
Classification
supervised, 62–3, 67, 73–4, 125
unsupervised, 59, 72–3, 102, 110
Cluster analysis, 59, 263
Clustering, 62, 92, 104, 110–11, 120–2, 149, 153–4, 163, 262, 264
Coating, 73, 75, 114, 124, 128, 251–2
Compartmental absorption and transport (CAT), 178
Competitive learning, 96, 103, 110, 148–9, 163
Compression, 3, 50, 52, 73–4, 112–13, 115–16, 119, 132
Computational fluid dynamics (CFD), 234–6, 240–56
Contour plot, 12, 26, 46
Convergence, 96, 142, 151, 153
Copied layer, 107
Correlation coefficient, 20, 63, 66, 71, 74, 109, 128, 134, 195, 211
Critical process parameters (CPP), 2–3, 6, 10–12, 46–8, 78
Critical quality attributes (CQA), 2–3, 5–6, 9–13, 47, 78
Crystallization, 143
Cubic design, 42, 53

D

D-optimal design, 19, 36, 40–2, 44, 51–2
Decision tree, 124–34, 136–9, 262–3
Degree of membership, 120–1
Delta rule, 98
Dendrogram, 62
Design space, 3–6, 8–9, 11–13, 18, 32, 198
Desirability function, 13, 52, 143
Dimensionality reduction, 60, 66, 68, 92, 111, 148
Disintegration (of tablets), 76, 112, 132–5
Doehlert design, 13, 39–40, 52
Dynamic networks, 96, 101, 104–7, 115–16

E

Elman (dynamic) neural network (EDNN), 106–7, 116
Emulsions
double (multiple), 18, 20–2
microemulsion, 22–7, 112–13
nanoemulsion, 114
o/w, 18–20, 22, 24, 112, 122
self-(micro)emulsifying system, 18, 22–6
Epoch, 96–8
Evolutionary computing, 139, 142
Experimental design (design of experiments, DOE), 5, 12–13, 19–22, 31–4, 40–2, 44, 46–51, 53–4, 58–9, 65, 69, 77–8, 81, 110–11, 143, 261

F

Factor
analysis (FA), 59, 64, 68
effects, 13, 19, 32–4, 36, 45, 47–8, 50, 52–4, 64, 77, 81
interactions, 11, 34–5, 42, 47, 49, 51, 53–4, 65, 77
Factorial design, 19, 34–8, 40, 46, 48–9, 202, 261
Fluid, 233–4, 236, 248
Fluid bed (processing), 136, 158, 251
Feature
function, 102
selection, 68, 111
space, 102, 163
Feed-forward, 6, 23, 26–7, 65, 96, 98–9, 101, 105–6
Fitness function, 140–1
Formulation composition, 18, 113, 115–16, 144
Friability, 112, 117–18, 132
Fuzzy
modeling, 25, 27, 120–2, 261–2

G

Gain ratio, 126–7
Gamma memory, 105–6
Genetic
algorithm (GA), 24, 26, 70, 99, 110, 120, 122, 127, 139–45, 261–2
operators, 140–3, 145
programming (GP), 139, 147, 261–2
Generalization, 26, 92, 100, 104, 216
Generalized regression neural network (GRNN), 26, 69, 101–3, 112–13, 115–18
Gini index, 126–7

H

Homogenization, 19, 21
Hydrogel, 112, 144–6, 156

I

ID3 algorithm 125–7,
In vitro-In vivo correlation (IVIVC), 52, 115–17, 183, 195, 210–11, 213, 216–17, 219, 221, 225, 246
Information gain, 126–7, 136
Inhaler, 117, 241–5
Interpolation weight, 105–6
Intrinsic solubility, 112, 186
Iteration, 96–8, 110, 149–51

J

J48 decision tree, 130–1

K

K-means clustering, 62, 102, 153
K-nearest neighbors (KNN), 63, 73
Kohonen networks, 62, 102, 104, 148–51, 153, 160, 163

L

Latent variables (LV), 60–3, 66–8, 75–8, 154–5
Layer (including input, output and hidden), 23–7, 68, 93–109
Learning rate, 98, 108, 151, 153
Least squares regression, 45, 48, 64–5, 67, 74, 97
Light scattering method, 118, 136–8, 143
Linear discriminant analysis (LDA), 62–3, 73
Liposomes, 12, 49, 130, 132
Loadings, 60–2, 67, 72–3, 81–3

M

Mean dissolution time (MDT), 113, 116, 123
Membership function, 121
Microparticles, 113
Mixing, 19, 76, 158–60, 247–8, 251
Mixture design, 19, 32–3, 42–4, 46
Modular neural network (MNN), 104, 118
Momentum factor, 98
Monte Carlo, 110
Multilayered perceptron (MLP), 24–6, 69, 99, 101, 104, 112–19, 123
Multivariate
classification, 59, 63
data analysis, 9, 12–13, 33, 49, 57–8, 69, 70–2, 75–8, 83, 144, 263–4
regression, 59, 63–4, 66–7, 134–5
spline interpolation (MSI), 21

N

NIPALS algorithm, 60

O

Objective function, 53, 143
Optimizer, 110, 139

P

Parameter sensitivity analysis (PSA), 198–203, 223, 225
Partial least squares (PLS), 46, 59, 63–4, 67–70, 73–6, 78, 81, 83, 85, 263
Partial least squares discriminant analysis (PLS-DA), 62–3
Pellets, 113, 116, 123, 252
Phase boundaries, 113
Plackett-Burman design, 34, 36, 48–50, 261
Preformulation, 112
Pretreatment methods, 70
Principal component
analysis (PCA), 13, 59, 61–4, 66–7, 69, 72–4, 76–8, 81, 83, 102, 153
regression (PCR), 59, 66–7, 69
Principal components (PC), 61, 73, 78, 81, 83, 152
Probabilistic neural networks, 63, 104
Probability density function, 102
Process analytical technology (PAT), 3, 5–6, 10, 12, 47, 58, 77
Pruning, 125, 127, 136

Q

Quality-by-design (QbD), 1–3, 5–14, 18, 46–7, 225
Quality risk
assessment, 4–6, 9, 12–13, 78
management, 2–5, 13
Quality target product profile (QTTP), 2–3, 5

R

Radial basis functions (RBF), 26, 68, 101, 103–4, 118, 142
Radial center, 102
Random forest, 127
Release rate, 13, 53, 115, 145–6, 154–6, 163, 183, 223–4
Regression
coefficients, 34, 45, 64–5, 70
model, 34, 42, 45, 64–5, 75–6, 134, 142, 147
multiple regression analysis (MRA), 19, 46, 59, 63–6, 70, 74
unit, 102
Residual analysis, 46, 50
Response surface, 20, 22, 26, 32–3, 36–7, 40, 42, 46, 52–4, 69, 93, 104, 164
Rheological properties, 20, 112
Robustness, 6, 32, 42, 44, 92, 142, 154
Root mean square error (RMSE), 71, 76, 220
Rules (induction), 120–5, 129, 132, 134, 136, 144

S

Scores, 61, 66–7, 76, 78, 81, 83–4
Screening design, 32–4, 36, 44, 46, 52
Self-organizing map (SOM), 63, 111, 148–64, 263
Signal recurrence, 96, 101, 106–7
Sigmoid function, 63, 68, 94–6, 101, 120, 141
Simplex lattice design, 42
Soft independent modeling of class analogy (SIMCA), 62–3, 73, 263
Splitting attribute, 125–6, 137–8
Spraying, 85, 113, 118, 136–9, 158–60, 251
Stability, 10, 18–22, 24, 49, 81, 83, 112, 115, 246
Standard error, 46, 71
Star design, 37
Static neural networks, 101–4
Support vector machine (SVM), 68–9, 73–4, 101–2, 143, 262

T

Tableting, 3, 47, 52, 75–6, 119
Tablets
coated, 73, 114, 124, 219, 252
controlled release (CR), 214–7
dispersible tablets, 12
extended release (ER), 11, 77, 115, 192–5, 201, 205, 208, 218
immediate release (IR), 2–3, 72–8, 81, 112–15, 119, 132–5, 143, 154, 186–95, 197, 201–2, 208, 211–22, 246
matrix, 52, 115–16, 146, 154, 156, 163
modified release (MR), 7, 47, 114
Taguchi design, 47
Taps, 105
Tensile strength, 51, 112, 116, 132–3, 135, 144
Time series, 105, 116
Topology (of ANN), 20, 23–7, 96, 104–5, 107–8, 110, 139, 148, 152
Transdermal drug delivery, 50–1, 117, 156

U

U-matrix, 152–3, 158

V

Validation (including cross-validation), 5–6, 8, 12, 22, 24, 26, 46, 59, 63, 70, 71, 76–7, 104, 108–9, 118, 127–8, 134, 136, 148, 183, 198, 241, 256
Variable importance on projection (VIP), 70
Virtual trial, 185, 204–5, 223–5

W

Weight vector, 60, 62, 67, 70
Weights, 68, 94–109, 149, 151