Note: ‘Page numbers followed by “f” indicate figures and “t” indicate tables.’
Adversarial networks,
185Alzheimer’s dementia
methodology for four-class characterization,
160comparison of methods,
163tworldwide development of,
158fAlzheimer’s Disease Neuroimaging Initiative dataset,
161Application programming interface (API),
100–101, 209Artificial learning,
131fAttention-based aspect extraction structure,
54, 55fBernsen’s binary method,
15–16Bidirectional long short-term memory (Bi-LSTM),
52–53distributed computing,
41–45velocity of data,
39, 39fvisualization of data,
41Biometric cryptosystems,
27–28Biometric recognition,
2–6, 26
expectations from biometric security system,
5–6identification
vs. verification,
5Biometric system, attack possibilities in,
25fsmear imaging diagnostic methods,
129Brazilian Public Health System,
201–202Cancelable biometrics,
26–27Casia.v4 iris database,
23Cell
checking and recognition,
252classification with ResNet-101,
257Classification and Regression Tree (CART),
229–230Classification stage,
269Cognitive normal (CN),
161Color
color-based approach,
266color-based image segmentation method,
271Common voice (Mozilla),
24Comparative analysis among different modalities,
28–29Computed tomography (CT),
143Computer-aided diagnostic (CAD),
143Concept-adapting Very Fast Decision Tree (CVFDT),
228–229Conceptual architecture,
164fConfirmatory data analysis (CDA),
37–38Conventional rule-based models,
119–120Convolutional neural network (CNN),
2–3, 5, 9f, 21–22, 51, 70–71, 76–77, 130, 139–141, 140f, 162, 170–172, 185, 190–193, 214–215, 216f, 242–250conceptual architecture,
164fconvolution operation,
10fgeneral architecture,
171fmonocyte classification,
150fmonocyte test image,
150fConvolutional neural network-fast (CNN-F),
16Creative Senz3D camera,
270CRoss Industry Standard Process for Data Mining (CRISP-DM),
50data mining stages in,
49f, 50Cross-sectional MRI Data in Young, Middle-Aged, Non-demented, and Demented Older Adults (OASIS-1),
170, 171tCross-validation (CV),
106CUDA deep neural network library (CuDNN),
187Curse of dimensionality,
51Customer relationship management (CRM),
56–59Cyber security
collection and preparation,
100–104data classification problem,
229–230multimedia data mining,
52–53Deep Boltzmann machine (DBM),
161Deep convolutional neural network,
160confusion matrix result of proposed architecture,
175foutcome of performance measures,
176tdeep learning methodology for dementia detection,
170–174autoencoder network,
6–28convolutional neural networks,
5, 9frecurrent neural networks,
5–6, 11fbasic functionality,
211fbiometric recognition,
3–6challenges in biometric recognition and security,
28characteristics of common modalities,
30tcomparative analysis among different modalities,
28–29multimedia data mining,
52–53methodology for precise recognition,
12–20methodology for spoof protection,
24–26methodology for template protection,
26–28biometric cryptosystems,
27–28cancelable biometrics,
26–27performance measures for intrusion detection systems,
218security aspects changing with,
219–223Deep learning-based detection and classification of adenocarcinoma cell nuclei
system architecture and methodology,
253–257Deep learning-based frameworks,
212–213Deep similarity learning model,
189measure of performance model,
203tin domain of deep learning and disease prediction,
191t–192tin domain of disease prediction,
194tDeep-stacked autoencoder,
161Degree of subjectivity,
90deep learning methodology for dementia detection,
170–174neuroimaging categorizing,
158research in molecular chemistry,
157–158worldwide development of Alzheimer’s dementia,
158fDenial of service (DOS),
209, 219Depth image datasets,
267Descriptive models,
50–51processing techniques,
66Digital smear diagnosis of blood smears,
129Discriminative models,
185Distributed computing,
41–45Distributed denial of service attack (DDOS attack),
209, 221–223Domain-specific (kidney-related) dataset,
193Drosophila melanogaster muscles,
142 Edge detection algorithm,
21–22Efficient Classification and Regression Tree algorithm (E-CART algorithm),
233–235, 234tcomputation time
comparison using hyperplane generator,
236tcomparison using SEA generator,
236tfor rotating hyperplane generator,
237fEfficient-Concept-adapting Very Fast Decision Tree (E-CVFDT),
228–229, 231–232Electronic Product Code technology (EPC technology),
121End-to-end deep learning,
2Equal error rate (EER),
6
deep learning methodology proposal,
145–147Exploratory data analysis (EDA),
37–38enrollment of facial biometric data,
15fidentification of facial biometric data,
14fFalse alarming rate (FAR),
218False match rate (FMR),
6False nonmatch rate (FNMR),
6Faster R-CNN algorithm, cell detection using,
255–257and selection for making feature vector,
268Feed-forward networks,
8–10Finger pore detection,
15–16Fingerprint recognition,
13–16Flow-based intrusion detection systems,
221Four
vs. of big data,
40, 40fsignature recognition,
22Gated recurrent unit (GRU),
161–162Gaussian decision tree (GDT),
233Gaussian skin color model,
266Generative adversarial networks,
217–218Google File System (GFS),
41Graphics processing unit (GPU),
3, 187Hadoop Distributed File System (HDFS),
41–42Hand gesture recognition studies,
265Handcrafted feature-driven basic machine learning classifiers,
97–98fingerprint recognition,
13–16palmprint recognition,
16–18sector
advantages and limitations of IoT,
124–125Hierarchical convolutional neural network (HCNN),
18–19for iris segmentation,
19fHigh Complexity Procedure Authorization (APAC),
201–202Histogram-based image segmentation,
268Histopathological image analysis,
120Hospital Admission Authorization (AIH),
201–202Hospital information system (HIS),
119–120Hypertext transfer protocol (HTTP),
123, 223in biometric recognition,
5Image preprocessing stage,
267layers configuration,
173tInductive logic programming,
73–74Information and communication technology (ICT),
114International Classification of Diseases (ICD-10),
200–201Internet of Medical Things,
113advantages and limitations for healthcare technology,
124–125discussions and future scope,
125–126security features for healthcare,
123–124Internet Protocol spoofing attack (IP spoofing attack),
209Intrusion detection systems,
219performance measures for,
218Iris Recognition Immigration System,
18
K-nearest neighbor classification,
229 Language recognition,
135Large-scale visualization,
38Learning
learning-based deep-Q networks,
219deep learning methodology proposal,
145–147Linear discriminant analysis,
54Local direction pattern (LDP),
21–22Lung adenocarcinoma (LUAD),
144–145Machine learning,
2, 51, 63, 65–66, 71–76, 81, 119, 135, 136f, 139, 159, 210–211, 251uses and applications of,
92–93Magnetic resonance imaging (MRI),
158Man-in-the-middle attack,
209, 210fMapping characteristics,
139Massively parallel processing,
41Mel frequency cepstral coefficient (MFCC),
20proposal modeling logic,
146fMicrosoft Cognitive Toolkit,
187, 188tMild cognitive impairment (MCI),
158Minutiae extraction network (MENet),
16similarity score calculation,
199fModern-day technology,
119Modified National Institute of Standards and Technology (MNIST),
76–77Multilayer perceptron (MLP),
187Multimedia data mining,
52–53Multimodal biometrics,
29Multiple-layer deep learning models,
120Multiscaled fully convoluted network (MFCN),
18–19National Laboratory of Pattern Recognition,
24Neuroimaging information,
159Neuropsychological information,
159NIGHT-care RFID system,
121One-shot enrollment,
12–13Open Access Series of Imaging Studies (OASIS),
170Open System Interconnection model,
222Palmprint recognition,
16–18Siamese network for palmprint recognition,
18fPassive attack phenomena,
207, 207fPhysiological parameters,
114, 122Point of vulnerabilities,
209comparison of different pooling operations results,
168fPositron emission tomography (PET),
158Precise recognition, deep learning methodology for,
12–20Pretrained AlexNet network,
253Proposed deep learning framework,
99, 100f
Radial basis function neural networks,
185Random neural networks,
219Real-time monitoring,
125Recurrent neural networks (RNNs),
2–3, 5–6, 11f, 21–22, 52, 95–96, 104, 161–162, 185–187, 214–215, 217Region proposal network (RPN),
256Regularized stacked denoising auto-encoder-softmax layer (RSDAE-SM),
190–193Reinforced learning,
73–74Reinforcement learning,
73ResNet-101, cell classification with,
257Resting-state functional magnetic resonance imaging (rs-fMRI),
160Restricted Boltzmann machine (RBM),
161, 185data collection and preparation,
100–104handcrafted feature-driven basic machine learning classifiers,
97–98proposed deep learning framework,
99, 100freceiver operating characteristic plot,
108frelated work and proposed work,
97–99research set-up and experimentation,
104–106Reverse bandwidth floods,
223aspects changing with deep learning,
219–223Segmentation stage,
268–269
feature extraction and selection for making feature vector,
268histogram-based image segmentation,
268thresholding methods,
268Self-organizing maps (SOMs),
51, 57–59Semisupervised learning,
73, 80Service attack, denial of,
220fSeven Vs of big data analysis,
37, 40SHA-512 cryptographic hashing,
12–13for palmprint recognition,
18fSign language recognition system,
266Signature recognition,
22Simple Service Discovery Protocol,
123Single neuron network,
94–95Social media platforms,
89Softmax
Softmax-based framework,
200Softmax-based supervised classification framework,
200Software and hardware configuration,
202Software-defined networking (SDN),
222Spam
Spatial constrained convolution neural network (SC-CNN),
260Squamous cell carcinoma of lung (LUSC),
144–145Squared error of regression line,
25Stacked denoising auto-encoder (SDAE),
190–193Stacked restricted Boltzmann machines,
220–221Stacked sparse autoencoder (SSAE),
260Structured information,
75Supervised techniques,
54Support vector machine (SVM),
119–120Apache Spark visualizations using,
45, 46f–47fTemplate protection,
26–28biometric cryptosystems,
27–28cancelable biometrics,
26–27Tensor processing unit (TPU),
3Tensor representation model,
77Thresholding methods,
268Transcription factor binding site (TFBS),
142Transcription factors (TFs),
141Transmission Control Protocol,
123Tumor development speed,
242Tweet
word cloud of most frequent terms in corpus,
103fTwo-phase strategy of feature fusion,
267Ubiris.v2 iris database,
23Unsupervised learning,
185, 197User Datagram Protocol,
123V1 (Job Tracker and Task Tracker),
41–42Verification in biometric recognition,
5Very Fast Decision Tree (VFDT),
228–229VGGNet-based algorithm,
15–16Viola–Jones algorithm,
12–13Visualization of data,
41preprocessing flowchart,
21fWang’s u-net model,
15–16Wireless sensor networks (WSNs),
121, 123based on Constrained Application Protocol,
122Zernike moment features,
269