Artificial Intelligence in Medicine: 11th Conference on Artificial Intelligence in Medicine, AIME 2007, Amsterdam, The Netherlands, July 7-11, 2007. Proceedings

Front matter......Page 1
Introduction......Page 15
State of the Art......Page 16
The Methodological Concepts......Page 17
A Multi-Agent System for Time Series Interpretation......Page 18
Agentification......Page 19
Rationale......Page 21
Time Series Data Exploration Case Study......Page 22
Discussion and Perspectives......Page 23
Introduction......Page 25
MRF Framework......Page 26
Local Approach of Tissue Segmentation......Page 27
Cooperative Approach of Tissue and Structure Segmentation......Page 28
Evaluation......Page 30
Discussion and Conclusion......Page 33
Introduction......Page 36
Background......Page 37
Focus Groups......Page 38
R-CAST......Page 39
R-CAST-MED: Adapting R-CAST to Emergency Medical Domain......Page 40
Simulation......Page 41
Supporting Decision Dependency and Multi-team Decision-Making......Page 43
Conclusion......Page 44
References......Page 45
The Maryland Virtual Patient Environment......Page 46
Multi-track Scripts: The Example of GERD......Page 48
Example: GERD with Erosive Esophagitis Leading to Erosion(s)......Page 51
Interactions and Interventions......Page 52
Patient Authoring......Page 53
Comparisons with Other Systems and Approaches......Page 54
References......Page 55
Introduction......Page 56
Agentification for the Analysis of Epileptic Signals......Page 57
Characterization of Connectivity Between Structures......Page 58
Results......Page 59
References......Page 60
Introduction......Page 61
HeCaSe2: A Distributed Guideline-Based Health Care System......Page 62
Ontological Representation of Medical Knowledge......Page 63
Conclusions......Page 64
Introduction......Page 66
Methods......Page 67
Results......Page 68
Determining Distinct Patient Groups in the Data Set: Clustering Analysis and Sub-group Discovery......Page 69
Determining If It Is Possible to Predict Changes in the Patient’s BP in Terms of Other Parameters......Page 72
Conclusions and Further Work......Page 74
References......Page 75
Introduction......Page 76
Reasoning and Temporal Abstraction by Means of FTCN......Page 78
Temporal Data Mining over FTCN......Page 80
Temporal Pattern Extension......Page 81
Conclusions and Future Works......Page 83
Introduction......Page 86
Cox Regression Models......Page 87
$k$-Nearest Neighbor Survival Prediction......Page 88
Data......Page 89
Design......Page 90
Results......Page 91
Discussion and Conclusions......Page 93
Introduction......Page 96
Preliminaries......Page 97
Dynamic Bayesian Networks......Page 98
Ventilator-Associated Pneumonia......Page 99
The NPC Algorithm......Page 100
Results......Page 101
Patients Not Diagnosed with VAP  ($D_{\\overline\\mathrm{vap}}$......Page 102
Patients With and Without VAP  ($D_{\\mathrm{VAP}}$......Page 103
Conclusions and Discussion......Page 104
Introduction......Page 106
Data and Methods......Page 107
Results......Page 110
Discussion and Related Work......Page 111
Introduction......Page 116
The Brain Ischaemia Data Analysis Problem......Page 117
Searching for Contrast Sets by Decision Tree Induction......Page 118
Contrast Set Mining Through Subgroup Discovery......Page 119
Round Robin Transformation: Unifying CSM and SD......Page 120
Implementations of Subgroup Discovery Algorithms and Subgroup Visualization in Orange......Page 121
Experimental Evaluation of the Round Robin CSM......Page 122
Experimental Evaluation of the One-Versus-All CSM......Page 123
Conclusions......Page 124
Introduction......Page 126
Stepwise Diagnostic Process......Page 127
Image Parametrization......Page 128
Medical Data......Page 130
Results......Page 131
Discussion......Page 134
Introduction......Page 137
Decision Trees......Page 138
Analogy to Statistical Hypothesis Tests......Page 140
Generating Alarm Rules......Page 141
Discussion......Page 144
Introduction......Page 146
Latent Variables......Page 148
Conclusion......Page 150
Introduction......Page 151
Generation of New Gene Sets......Page 152
Experiments and Conclusion......Page 153
Variable Selection for Decision Making......Page 156
Variable Ranking for Qualitative Interactions......Page 158
Nefazodone CBASP Trial......Page 159
Conclusion......Page 160
Introduction......Page 162
Statistical Approach......Page 163
Induction of Coexisting Factors......Page 165
Conclusions......Page 166
Introduction......Page 167
Knowledge Discovery in Medical Documents......Page 168
Association Rules in Query Refinement and Expansion......Page 170
References......Page 171
Introduction......Page 172
Learning the Predictive Model......Page 173
Results......Page 174
Conclusion and Future Work......Page 176
Introduction......Page 177
Definition of the Context......Page 178
Computation of the Training Set for Decision Tree Learning......Page 179
Experiments and Comparison with Expert Rules......Page 180
References......Page 181
Introduction......Page 182
Approach to Human Resources Monitoring......Page 183
Qualification of Physicians......Page 184
Short Description of Other Modules......Page 185
References......Page 186
Introduction......Page 187
OLAP Engine......Page 188
Results and Conclusion......Page 190
References......Page 191
Introduction......Page 192
Material......Page 193
Results and Discussion......Page 194
Conclusion and Perspectives......Page 195
Clustering Algorithm for Cancer Data......Page 197
Descriptive Analysis of the Population Cancer Mortality Data......Page 199
Concluding Remarks......Page 201
Introduction......Page 202
Classic and Novel Loop Features......Page 203
Classification and Results......Page 205
Conclusions......Page 206
Introduction......Page 207
Clustering Approach......Page 209
Interpretation and a Tentative Conclusion......Page 210
References......Page 211
Introduction......Page 212
The Palga Data......Page 213
Support Vector Machines......Page 214
Related Work......Page 215
Feature Engineering......Page 216
Changing Learning Parameters and Output Classes......Page 217
A Comparison with Domain Experts......Page 218
Concluding Remarks......Page 220
Introduction......Page 222
Method RaJoLink......Page 224
Search for Joint Terms in the Literature About Rare Terms......Page 225
Search for Linking Terms and the Corresponding Pairs of Articles......Page 226
Re-application of RaJoLink on a Restricted Domain......Page 227
Conclusions......Page 229
References......Page 230
Introduction......Page 232
Input Data......Page 233
Architecture......Page 234
Signal Analysis......Page 235
Data Abstraction......Page 236
Content Determination......Page 237
Preliminary Results......Page 238
Discussion......Page 240
References......Page 241
Anonymisation......Page 242
Method......Page 243
A Corpus of Clinical Data and Evaluation......Page 245
References......Page 246
Overview......Page 247
Preliminary Tasks......Page 248
Method......Page 249
Results......Page 250
Conclusion......Page 251
Introduction......Page 252
The Matching Technique......Page 253
Experiments......Page 254
References......Page 256
Introduction......Page 257
Material and Method......Page 258
Results and Discussion......Page 259
Conclusion and Perspectives......Page 260
Introduction......Page 262
Textpreprocessor......Page 263
Semantic Interpreter......Page 264
Extractor......Page 265
Results......Page 266
Discussion......Page 267
Conclusion......Page 268
References......Page 269
Introduction......Page 270
OWL and SWRL......Page 271
OWL-Relational Mapping......Page 272
Knowledge-Level Querying with SWRL......Page 273
Schema and Mapping Ontologies......Page 275
Mapping Software and Query Engine......Page 276
References......Page 278
Introduction......Page 280
Name Changes in Fungal Taxonomy......Page 281
Managing Name Changes......Page 282
Category Theory and Ontologies......Page 283
The Category Class......Page 284
Managing Changes Using Category Theory......Page 285
Application Scenario......Page 287
References......Page 288
Introduction......Page 290
SEP-Triplets in SNOMED CT......Page 291
Replacing SEP-Triplets by Using the DL  $\\EL^+$......Page 292
Medical Ontologies Modeling......Page 295
Material and Methods......Page 296
Bottom-Up Steps: Considering Free-Text Answers in Medical Records......Page 297
The Need for a Customized Modeling Approach......Page 298
Future Work......Page 299
Introduction......Page 300
Analyzing Differences in Definitions......Page 301
Using the Operationalization Hierarchy in Comparing Definitions......Page 302
Discussion......Page 303
Introduction......Page 306
Patients......Page 308
Resource Calendar......Page 309
Simulation......Page 310
Adaptive Model......Page 312
Experiments......Page 314
Conclusions......Page 315
Introduction......Page 317
Information Measures and Test Selection......Page 318
The Measures from a Fundamental Perspective......Page 320
The Experimental Results......Page 323
Conclusions......Page 326
Introduction......Page 327
Application to NPC Data......Page 328
Conclusion......Page 330
Introduction......Page 332
Preliminaries......Page 333
Enhanced Test-Selection Algorithms......Page 334
Conclusions......Page 336
Introduction......Page 337
System Description......Page 338
Discussion and Future Work......Page 340
Introduction......Page 342
AQUA and Its Web Content Collection Subsystem......Page 343
Evaluation Methodology and Results......Page 344
Conclusions and Future Work......Page 345
References......Page 346
Introduction......Page 347
Bayesian Networks......Page 348
Structure Representation of the Breast Cancer Domain......Page 349
Conclusion......Page 351
Introduction......Page 352
Automated Extraction of Optimal Pathways......Page 353
Model Structure......Page 354
Automated Representation of the Optimal Decision Paths in a Tree-Shaped Flow Diagram......Page 355
Diagnosing Acute Chest Pain......Page 356
References......Page 358
Introduction......Page 360
Intrinsic Factors......Page 361
Plausibility Model......Page 362
Conclusion......Page 364
Introduction......Page 365
Background: Computer-Based Tools......Page 366
An Easier Interface for E-Prime......Page 368
The Image Elaboration Tool......Page 369
Novelty of the Approach......Page 371
Results......Page 373
References......Page 374
Introduction......Page 376
Acquisition and Preprocessing of Images......Page 377
Footprint Segmentation Using Active Contours......Page 378
Footprint Representation and Characteristics Extraction......Page 379
Training of the Neuronal Network......Page 380
Classification of the Training Set......Page 381
Classification of the Validation Set......Page 382
Conclusions......Page 383
Introduction......Page 386
Experimentation and Analyses......Page 387
References......Page 390
Introduction......Page 391
Functional Canonical Correlation......Page 392
Brain State Prediction......Page 393
Results......Page 394
Conclusion......Page 395
Introduction......Page 396
Model-Centric Approaches......Page 397
LASSIE – Modeling Treatment Processes Using Information Extraction......Page 398
Adaptation of LASSIE for ’Living Guidelines’......Page 399
Marking-Up the New Guideline Version......Page 400
Further Transformation of the Extracted Information......Page 401
Formalizing the Original Guideline Version......Page 402
Discussion......Page 403
References......Page 404
Introduction......Page 406
Approach......Page 407
Preliminaries......Page 408
Critiquing Formulas......Page 410
Case Study 1: Ductal Carcinoma in Situ......Page 411
Case Study 2: Infiltrating Ductal Carcinoma......Page 412
Related Work......Page 413
Conclusions......Page 414
Introduction and Rationale......Page 416
Modeling Guidelines with Planning Formalisms......Page 417
System Architecture and Overview......Page 418
Document Processing Applied to Clinical Guidelines......Page 419
Synchronization of HTN Traversal and Document Exploration......Page 420
Example Results: The French Hypertension Guidelines^1......Page 422
Discussion......Page 423
References......Page 424
Introduction......Page 426
Problem Description......Page 427
Method......Page 428
Results......Page 430
Discussion......Page 434
References......Page 435
Introduction......Page 436
The Models......Page 437
Modifying the Models......Page 438
Conclusion......Page 439
Introduction and Background......Page 441
Framework for Clinical Practice Guidelines......Page 442
Learn the Causal Model from Practice Data and Recognize the Likely CPGs Being Followed......Page 443
Discussion and Conclusion......Page 444
Introduction......Page 446
CPG Modeling Module......Page 448
Execution Sub-module......Page 449
References......Page 450
Introduction......Page 451
Inference in the Promedas Graphical Model......Page 452
Simulations with Virtual Patient Data......Page 453
Conclusions......Page 454
Introduction......Page 456
Problems Encountered During Implementation......Page 457
The Problems Capturing Module......Page 458
Results......Page 459
References......Page 460
Introduction......Page 462
A Motivating Problem......Page 464
A Similarity Proposal for Clinical Cases......Page 466
Discussion and Conclusions......Page 470
Introduction......Page 472
A Brief Description of the SCIFF Framework......Page 473
Translation Algorithm......Page 475
Mapping of a Minimal Window to an $IC$......Page 476
General Algorithm......Page 477
A Case Study......Page 478
Conclusions......Page 480
Introduction......Page 482
Detecting Disease Conditions......Page 484
Using State-Transition Diagrams to Represent the Cases in Hospital DBs......Page 486
Induction of Partial Orders......Page 487
The Statistical Model......Page 488
Experiments......Page 489
Results on the Condition-Based Prognosis......Page 490
References......Page 491
Introduction......Page 493
Architecture......Page 494
Application......Page 495
Conclusion......Page 497
Back matter......Page 499