Medical Data Sharing, Harmonization and Analytics

Cover......Page 1
Medical Data Sharing, Harmonization and Analytics
......Page 2
Copyright......Page 3
Preface......Page 4
9.1 Conclusions......Page 346
List of abbreviations......Page 9
1.1 Origin of medical data......Page 14
1.2 Toward medical data sharing and harmonization......Page 16
1.3 Distributed data processing architectures......Page 20
1.4 Scope and contribution to the state of the art......Page 23
1.5 Organizational structure......Page 26
References......Page 30
2 - Types and sources of medical and other related data......Page 32
2.1 Overview......Page 33
2.2.1 Biosignals......Page 36
2.2.2 Medical images......Page 38
2.2.3 Omics......Page 42
2.2.4 Laboratory tests......Page 45
2.3.1 Biosignal acquisition standards......Page 47
2.3.2 Laboratory tests standards......Page 48
2.3.3 Medical imaging acquisition standards......Page 50
2.3.4 Omics acquisition standards......Page 52
2.4.2 Health sensors......Page 54
2.4.4 Genome registries......Page 55
2.4.5 Clinical trials......Page 56
2.5.1 Origins......Page 57
2.5.2 Cohort study design......Page 58
2.5.3 Comparison with other study designs......Page 60
2.6 Big data in medicine......Page 61
2.7 Conclusions......Page 64
References......Page 69
3.1 Overview......Page 79
3.2 The rationale behind medical data sharing......Page 81
3.2.1 Patient stratification......Page 82
3.2.2 Identification of new biomarkers and/or validation of existing ones......Page 83
3.2.3 New therapy treatments......Page 84
3.3 Data curation......Page 85
3.3.2 Data annotation......Page 86
3.3.4 Data imputation......Page 88
3.3.5 Outlier detection......Page 89
3.4 Standardization......Page 96
3.5.1 Framework for responsible sharing for genomic and health-related data......Page 99
3.5.2 The DataSHIELD framework......Page 101
3.6.1 ClinicalTrials.gov......Page 103
3.6.2 The database for Genotypes and Phenotypes......Page 104
3.6.4 Biogrid Australia......Page 106
7.6.8 Amazon Web Services machine learning......Page 107
3.6.6 The Query Health initiative......Page 108
3.7 Solutions against the misuse of clinical data......Page 109
3.8 Conclusions......Page 111
References......Page 113
4.1 Overview......Page 117
4.2 The fundamental basis of data governance......Page 120
4.3.2 Patient privacy issues......Page 122
4.3.3 Technical limitations......Page 123
4.3.4 Other aspects......Page 124
4.4.1 The Directive 95/46/EC of the European Parliament and of the Council......Page 125
4.4.2 The General Data Protection Regulation......Page 127
4.4.3 The Lisbon Treaty and its impact in data protection law development......Page 130
4.5.1 The Federal Trade Commission Act......Page 131
4.5.2 The Health Insurance Portability and Accountability Act......Page 133
7.6.2 Scikit-learn......Page 135
4.6 Overlapping between EU and US protection laws......Page 136
4.7 Global initiatives......Page 139
4.8 Toward a more complete data protection framework......Page 141
4.9 Conclusions......Page 143
References......Page 146
5.1 Overview......Page 149
5.2 The origins and prospects of harmonizing datasets......Page 153
5.3 Cohort integration requirements......Page 154
5.4 Barriers toward medical data harmonization......Page 155
5.5.1 The stringent approach......Page 156
5.5.2 The flexible approach......Page 157
6.6 Security protocols and guidelines......Page 159
5.6.2 Semantic matching......Page 165
5.6.3.1 Item response theory......Page 172
5.6.3.2 Linear factor and multiple factor analysis......Page 177
5.6.3.3 Generalized linear factor analysis......Page 180
5.6.3.4 Moderated nonlinear factor analysis......Page 181
5.7 Existing frameworks......Page 185
5.8 Toward a generalized harmonization strategy......Page 187
5.9 Conclusions......Page 189
References......Page 191
6.1 Overview......Page 196
6.2 The concept of cloud computing......Page 199
6.3 Web services......Page 205
6.4 Architectures......Page 210
6.4.1 Infrastructure as a service......Page 212
6.4.2 Platform as a service......Page 214
6.4.4 Data as a service......Page 215
6.5 Storage......Page 216
6.6.1 Cloud security alliance......Page 220
6.6.2 Institute of electrical and electronic engineers standards......Page 222
6.6.3 European network and information and security agency guidelines......Page 223
6.6.4 National institute of standards and technology guidelines......Page 225
6.7 Challenges......Page 227
6.8 Conclusions......Page 230
References......Page 232
Further reading......Page 237
7 - Machine learning and data analytics......Page 238
7.1 Overview......Page 239
7.2.2 Data discretization......Page 242
8.2.3 Phenotypes and epidemics......Page 326
7.3.2.1.1 Linear and multiple regression......Page 249
7.3.2.1.2 Logistic regression......Page 252
7.3.2.2 Support vector machines......Page 254
7.3.2.3 Naïve Bayes......Page 260
7.3.2.4.1 Classification and Regression Trees......Page 263
7.3.2.4.2 ID3, C4.5, and C5.0 implementations......Page 265
7.3.2.4.3 Ensemble classifiers......Page 269
7.3.3 Artificial neural networks and deep learning......Page 275
7.3.4 Performance evaluation......Page 284
7.3.5.1 K-means......Page 287
7.3.5.2 Spectral clustering......Page 289
7.3.5.3 Hierarchical clustering......Page 291
7.3.5.4 Other data clustering approaches......Page 292
7.3.5.5 Clustering performance evaluation......Page 294
7.4 Distributed learning......Page 297
7.5 Visual analytics......Page 299
7.6.1 Apache Spark......Page 302
7.6.5 Weka......Page 305
7.7 Examples of applications in the medical domain......Page 306
7.8 Conclusions......Page 307
References......Page 312
8.1 Overview......Page 321
8.2.1 Aging studies......Page 322
8.2.2 Obesity......Page 325
8.2.5 Personality scores......Page 329
8.2.6 Other case studies......Page 330
8.2.7 Ongoing projects......Page 331
8.3 Conclusions......Page 333
References......Page 343
9.2 Future trends......Page 350
References......Page 353
C......Page 354
D......Page 356
E......Page 358
G......Page 359
H......Page 360
J......Page 361
M......Page 362
N......Page 363
P......Page 364
R......Page 365
S......Page 366
T......Page 367
W......Page 368
X......Page 369
Back Cover......Page 370