<p>List of Contributors xiii</p> <p>Preface xix</p> <p><b>1 Introduction 1<br /> </b><i>Bruno Carpentieri and Paola Lecca</i></p> <p>1.1 Disease Diagnoses 4</p> <p>1.2 Drug Development 6</p> <p>1.3 Personalized Medicine 6</p> <p>1.4 Gene Editing 7</p> <p>Author Biographies 9</p> <p>References 9</p> <p><b>2 Fuzzy Logic for Knowledge-Driven and Data-Driven Modeling in Biomedical Sciences 17<br /> </b><i>Paolo Cazzaniga, Simone Spolaor, Caro Fuchs, Marco S. Nobile and Daniela Besozzi</i></p> <p>2.1 Introduction 17</p> <p>2.2 Fuzzy Logic 18</p> <p>2.2.1 Fuzzy Sets 19</p> <p>2.2.2 Linguistic Variables 19</p> <p>2.2.3 Fuzzy Rules 20</p> <p>2.2.4 Fuzzy Inference Systems 21</p> <p>2.2.5 Simpful 22</p> <p>2.3 Knowledge-Driven Modeling 22</p> <p>2.3.1 Dynamic Fuzzy Modeling 23</p> <p>2.3.2 Application 1: Maximizing Cancer Cells Death with Minimal Drug Combinations 25</p> <p>2.3.3 FuzzX: A Hybrid Mechanistic-Fuzzy Modeling and Simulation Engine 27</p> <p>2.3.4 Application 2: Analyzing Oscillatory Regimes in Signal Transduction Pathways 29</p> <p>2.4 Data-Driven Modeling 30</p> <p>2.4.1 pyFUME: Automatic Generation of Fuzzy Inference Systems 31</p> <p>2.4.2 Application 3: Assessing Tremor Severity in Neurological Disorders 33</p> <p>2.5 Discussion 35</p> <p>Author Biographies 36</p> <p>References 37</p> <p><b>3 Application of Machine Learning Algorithms to Diagnosis and Prognosis of Chronic Wounds 43<br /></b><i>Mai Dabas and Amit Gefen</i></p> <p>3.1 Background 43</p> <p>3.1.1 Chronic Wounds 43</p> <p>3.1.2 Implementation of AI Methodologies in Wound Care and Management 43</p> <p>3.2 Clinical Visual Assessment of Wounds Supported by Artificial Intelligence 44</p> <p>3.2.1 Predicting the Formation and Progress of Wounds Based on Electronic Health Records 46</p> <p>3.2.2 Predicting the Formation and Evolution of Wounds Based on a Dynamic Evaluation of Wound Characteristics and Relevant Physiological Measures 48</p> <p>3.2.3 Feasible Implementation of AI Solutions For Wound Care Delivery and Management 49</p> <p>3.2.4 Types of Data Modalities for Diagnosis, Detection, and Prediction of Chronic Wounds 50</p> <p>3.3 Smartphone and Tablet Use in Wound Diagnosis and Management 51</p> <p>3.4 Conclusions 53</p> <p>Acronyms 54</p> <p>Author Biographies 55</p> <p>References 55</p> <p><b>4 Deep Learning Techniques for Gene Identification in Cancer Prevention 59<br /> </b><i>Eleonora Lusito</i></p> <p>4.1 The Next-Generation Era of Cancer Investigation 59</p> <p>4.1.1 Cancer at Its First Definitions 59</p> <p>4.1.2 Attempts to Sequence Nucleic Acids Over the Years 60</p> <p>4.1.3 From the First to the Third-Generation Sequencing 61</p> <p>4.1.4 Applications of NGS in Clinical Oncology 62</p> <p>4.2 Deep Learning Approaches for Genomic Variants Identification in Cancer 63</p> <p>4.2.1 Cancer Causing Factors 63</p> <p>4.2.2 The Contribution of Germline Alterations to Cancer 64</p> <p>4.2.3 Somatic Mutations and Cancer 64</p> <p>4.2.4 Calling Variants from Sequence Data 65</p> <p>4.2.5 Computational Approaches for Variant Discovery 65</p> <p>4.2.6 Convolutional Neural Networks (CNNs): Basic Principles 66</p> <p>4.2.7 Application of CNNs to Variant Calling 67</p> <p>4.2.8 A Typical CNN Architecture for Variant Calling 68</p> <p>4.2.9 The Activation Function 69</p> <p>4.2.10 Dropout and L1–L2 Regularization 71</p> <p>4.2.11 Advantages of Deep Learning Over the Existing Techniques 72</p> <p>4.2.12 Residual Neural Networks (ResNet)-Inspired CNN in Genomic Variants Detection 73</p> <p>4.3 Deep Learning in Cancer Transcriptomics 74</p> <p>4.3.1 Gene Expression and Cancer 74</p> <p>4.3.2 Analytical Approaches to Deal with Gene Expression Data 76</p> <p>4.3.3 Stacked Denoising Autoencoders (SDAEs) for Dimensionality Reduction 76</p> <p>4.3.4 The Variational Autoencoder (VAE) 79</p> <p>4.3.5 VAEs to Integrate Gene Expression and Methylation Data 81</p> <p>4.3.5.1 DNA Methylation: the Epigenetic Regulation of Gene Expression 81</p> <p>4.3.5.2 Preprocessing Input Data of Different Sources 82</p> <p>4.3.5.3 A VAE Architecture for Multimodal Data 82</p> <p>4.4 Conclusions 84</p> <p>Acronyms 86</p> <p>Author Biographies 87</p> <p>References 87</p> <p><b>5 Deep Learning for Network Biology 97<br /> </b><i>Eleonora Lusito</i></p> <p>5.1 Types of Interactions Between Genes and Their Products 97</p> <p>5.2 Deep Learning Methods with Graph-input Data 99</p> <p>5.2.1 Graph Embedding 99</p> <p>5.2.1.1 Random Walk-Based Graph Embedding 100</p> <p>5.2.1.2 Proximity-Based Graph Embedding 101</p> <p>5.2.2 Graph Convolutional Networks (GCNs) 102</p> <p>5.3 Applications of GNNs to Infer Biological and Pharmacological Interactions 104</p> <p>5.3.1 Proteomics 104</p> <p>5.3.2 Drug Development and Repurposing 104</p> <p>5.3.3 Drug–Drug Interaction Prediction 105</p> <p>5.3.4 Disease Classification and Outcome Prediction 106</p> <p>Author Biography 107</p> <p>References 107</p> <p><b>6 Deep Learning-Based Reduced Order Models for Cardiac Electrophysiology 115<br /> </b><i>Stefania Fresca, Luca Dedè and Andrea Manzoni</i></p> <p>6.1 Overview of Cardiac Physiology 115</p> <p>6.1.1 Atrial Tachycardia and Atrial Fibrillation 117</p> <p>6.1.2 Mathematical Models for Cardiac Electrophysiology 118</p> <p>6.2 Reduced Order Modeling 121</p> <p>6.2.1 Problem Formulation 123</p> <p>6.2.2 Nonlinear Dimensionality Reduction 123</p> <p>6.3 Decreasing Complexity in Cardiac Electrophysiology 124</p> <p>6.3.1 POD-Enhanced Deep Learning-Based ROMs 125</p> <p>6.3.1.1 POD-DL-ROM Architecture and Algorithms 128</p> <p>6.4 Numerical Results 130</p> <p>6.4.1 Test 1: Two-Dimensional Slab with Figure of Eight Reentry 131</p> <p>6.4.2 Test 2: Three-Dimensional Left Ventricle Geometry 133</p> <p>6.4.3 Test 3: Left Atrium Surface by Varying the Stimuli Location 135</p> <p>6.4.4 Test 4: Reentry Breakup 137</p> <p>6.5 Conclusions 139</p> <p>Author Biographies 140</p> <p>References 140</p> <p><b>7 The Potential of Microbiome Big Data in Precision Medicine: Predicting Outcomes Through Machine Learning 149<br /> </b><i>Silvia Turroni and Simone Rampelli</i></p> <p>7.1 The Gut Microbiome: A Major Player in Human Physiology and Pathophysiology 149</p> <p>7.2 Machine Learning Applied to Microbiome Research 151</p> <p>7.2.1 Case Study 1: Obesity 151</p> <p>7.2.2 Case Study 2: Cancer 153</p> <p>7.2.3 Case Study 3: Personalized Nutrition 154</p> <p>7.2.4 Case Study 4: Exploiting the Meta-Community Theory for New Machine Learning Approaches 155</p> <p>7.3 Conclusions and Perspectives 155</p> <p>Author Biographies 156</p> <p>References 156</p> <p><b>8 Predictive Patient Stratification Using Artificial Intelligence and Machine Learning 161<br /> </b><i>Thanh-Phuong Nguyen, Thanh T. Giang, Quang T. Pham and Dang H. Tran</i></p> <p>8.1 Overview of Artificial Intelligence for Patient Stratification 161</p> <p>8.2 A RPCA and MKL Combination Model for Patient Stratification 164</p> <p>8.2.1 Robust Principal Component Analysis 164</p> <p>8.2.2 Dimensionality Reduction and Features Extraction Based on RPCA 166</p> <p>8.2.3 Predictive Model Construction Based on Multiple Kernel Learning 168</p> <p>8.2.4 Materials 169</p> <p>8.2.4.1 Cancer Patient Datasets 169</p> <p>8.2.4.2 Alzheimer Disease Patient Datasets 170</p> <p>8.2.5 Experiment Design 171</p> <p>8.2.5.1 Experiment of Stratifying Cancer Patients 171</p> <p>8.2.5.2 Experiment of Stratifying Alzheimer Disease Patients 171</p> <p>8.2.6 Results and Discussions 171</p> <p>8.2.6.1 Application of Stratifying Cancer Patients 172</p> <p>8.2.7 Application of Stratifying Alzheimer Disease Patients 174</p> <p>8.3 Conclusion 175</p> <p>Author Biographies 175</p> <p>References 176</p> <p><b>9 Hybrid Data-Driven and Numerical Modeling of Articular Cartilage 181<br /> </b><i>Seyed Shayan Sajjadinia, Bruno Carpentieri and Gerhard A. Holzapfel</i></p> <p>9.1 Introduction 181</p> <p>9.2 Knee and Cartilage 182</p> <p>9.2.1 Main Joint Substructures 182</p> <p>9.2.2 Load-Bearing Cartilage Phases 183</p> <p>9.3 Physics-Based Modeling 185</p> <p>9.3.1 Numerical Modeling 185</p> <p>9.3.2 Constitutive Modeling 188</p> <p>9.4 AI-Enhanced Modeling 191</p> <p>9.4.1 Deep Learning 191</p> <p>9.4.2 Surrogate Modeling 192</p> <p>9.5 Discussion and Conclusion 194</p> <p>Author Biographies 194</p> <p>References 195</p> <p><b>10 A Hybrid of Differential Evolution and Minimization of Metabolic Adjustment for Succinic and Ethanol Production 205<br /> </b><i>Zhang N. Hor, Mohd S. Mohamad, Yee W. Choon, Muhammad A. Remli and Hairudin A. Majid</i></p> <p>10.1 Introduction 205</p> <p>10.2 Method 206</p> <p>10.2.1 Differential Evolution (DE) 206</p> <p>10.2.2 Mutation 206</p> <p>10.2.3 Crossover 207</p> <p>10.2.4 Selection 208</p> <p>10.2.5 Minimization of Metabolic Adjustment 208</p> <p>10.2.6 A Hybrid of Differential Evolution and Minimization of Metabolic Adjustment 209</p> <p>10.3 Experiments and Discussion 209</p> <p>10.3.1 Dataset 209</p> <p>10.3.2 Parameter Setting 209</p> <p>10.3.3 Experimental Results 210</p> <p>10.3.4 Comparative Analysis 214</p> <p>10.4 Conclusion 214</p> <p>Acknowledgment 215</p> <p>Author Bibliographies 215</p> <p>References 216</p> <p><b>11 Analysis Pipelines and a Platform Solution for Next-Generation Sequencing Data 219<br /> </b><i>Víctor Duarte, Alesandro Gómez and Juan M. Corchado</i></p> <p>11.1 Introduction 219</p> <p>11.2 NGS Data Analysis Pipeline and State of the Art Tools 220</p> <p>11.2.1 Quality Assessment 220</p> <p>11.2.2 Alignment 221</p> <p>11.2.3 Post-alignment and pre-variant Calling Processing 222</p> <p>11.2.4 Variant Calling 223</p> <p>11.2.5 Variant Annotation 228</p> <p>11.3 Nanopore Sequencing Data Analysis 229</p> <p>11.3.1 Base-Calling 230</p> <p>11.3.2 Quality Control and Preprocessing 230</p> <p>11.3.3 Error Correction 231</p> <p>11.3.4 Alignment 231</p> <p>11.3.5 Variant Calling 231</p> <p>11.4 Machine Learning Approaches in Variant Calling 232</p> <p>11.5 Next-Generation Sequencing Data Analysis Frameworks 233</p> <p>11.6 DeepNGS 235</p> <p>11.6.1 Pipeline 235</p> <p>11.6.2 DeepNGS Main Features 236</p> <p>11.6.2.1 Power and Speed 236</p> <p>11.6.2.2 Optimized Workflow 236</p> <p>11.6.2.3 Intuitive Design and Interactive Charts 237</p> <p>11.6.2.4 Extended Information 237</p> <p>11.6.2.5 Artificial Intelligence and Machine Learning 237</p> <p>11.7 Conclusions 240</p> <p>Author Biographies 241</p> <p>References 241</p> <p><b>12 Artificial Intelligence: From Drug Discovery to Clinical Pharmacology 253<br /> </b><i>Paola Lecca</i></p> <p>12.1 Artificial Intelligence and the Druggable Genome 253</p> <p>12.2 Feature-Based Methods 257</p> <p>12.3 Similarity/Distance-Based Methods 257</p> <p>12.4 Matrix Factorization 258</p> <p>12.4.1 Causal K-Nearest-Neighborhood 261</p> <p>12.4.2 Causal Random Forests 263</p> <p>12.4.3 Causal Support Vector Machine 264</p> <p>12.5 Opportunities and Challenges 265</p> <p>Author Biography 266</p> <p>References 266</p> <p><b>13 Using AI to Steer Brain Regeneration: The Enhanced Regenerative Medicine Paradigm 273<br /> </b><i>Gabriella Panuccio, Narayan P. Subramaniyam, Angel Canal-Alonso, Juan M. Corchado and Carlo Ierna</i></p> <p>13.1 The Challenge of Brain Regeneration 273</p> <p>13.2 The Enhanced Regenerative Medicine Paradigm 274</p> <p>13.3 The Case of Epilepsy 276</p> <p>13.4 AI to Understand Epilepsy 279</p> <p>13.4.1 Commonly Applied Learning Algorithms for Basic Neuroscience and Clinical Application in Epilepsy 282</p> <p>13.4.2 Seizure and Epilepsy Type Classification 284</p> <p>13.4.3 Seizure Onset Zone Localization 284</p> <p>13.4.4 Seizure Detection 285</p> <p>13.4.5 Seizure Prediction 285</p> <p>13.4.6 Signal Feature Extraction for Seizure Detection and Prediction 288</p> <p>13.4.7 Network Interactions and Evolving Dynamics in the Epileptic Brain: The Eye of AI 290</p> <p>13.5 Artificial Intelligence to Guide Graft-Host Dynamics in Epilepsy 292</p> <p>13.6 Challenges and Limitations 294</p> <p>13.6.1 From AI to Explainable AI 295</p> <p>13.7 A Philosophical Perspective on Enhanced Brain Regeneration 297</p> <p>Acknowledgments 299</p> <p>Acronyms 299</p> <p>Author Biographies 300</p> <p>References 300</p> <p><b>14 Towards Better Ways to Assess Predictive Computing in Medicine: On Reliability, Robustness, and Utility 309<br /> </b><i>Federico Cabitza and Andrea Campagner</i></p> <p>14.1 Introduction 309</p> <p>14.2 On Ground Truth Reliability 311</p> <p>14.2.1 Weighted Reliability 314</p> <p>14.2.2 Example Application 316</p> <p>14.3 On Utility Metrics to Evaluate ML Performance 318</p> <p>14.3.1 Weighted Utility 318</p> <p>14.3.2 Example Application 321</p> <p>14.4 On the Replicability of Clinical ML Models 322</p> <p>14.4.1 Dataset Size 323</p> <p>14.4.2 Dataset Similarity 325</p> <p>14.4.3 Meta-Validation Procedure 325</p> <p>14.4.4 Example Application 328</p> <p>14.5 Conclusions and Future Outlook 331</p> <p>Author Biographies 332</p> <p>References 333</p> <p><b>15 Legal Aspects of AI in the Biomedical Field. The Role of Interpretable Models 339<br /> </b><i>Chiara Gallese</i></p> <p>15.1 Introduction 339</p> <p>15.2 Data Protection 340</p> <p>15.3 Transparency Principle 343</p> <p>15.3.1 Right of Explanation 343</p> <p>15.3.2 Right of Information 348</p> <p>15.3.3 Informed Consent Requirements 349</p> <p>15.4 Accountability Principle 350</p> <p>15.5 Non-discrimination Principle and Biases 351</p> <p>15.6 High-Risk Systems and Human Oversight 353</p> <p>15.7 Additional Requirements of the AI Act Proposal 354</p> <p>15.8 Interpretability as a Standard 355</p> <p>15.9 Conclusion 358</p> <p>Author Biography 358</p> <p>References 359</p> <p><b>16 The Long Path to Usable AI 363<br /> </b><i>Barbara Di Camillo, Enrico Longato, Erica Tavazzi and Martina Vettoretti</i></p> <p>16.1 Promises and Challenges of Artificial Intelligence in Healthcare 363</p> <p>16.2 Deployment of Usable Artificial Intelligence Models 367</p> <p>16.2.1 Case Study: Predicting the Cardiovascular Complications of Diabetes via a Deep Learning Approach 368</p> <p>16.3 Potential and Challenges of Employing Longitudinal Clinical Data in AI 375</p> <p>16.3.1 Case Study: Modeling the Progression of Amyotrophic Lateral Sclerosis Through a Dynamic Bayesian Network 378</p> <p>16.3.2 Case Study: Investigating Amyotrophic Lateral Sclerosis Progression Trajectories Leveraging Process Mining 381</p> <p>16.4 Enhancing the Applicability of AI Predictive Models by a Combined Model Approach: A Case Study on T2D Onset Prediction 386</p> <p>16.4.1 The Problem of Type 2 Diabetes Prediction 386</p> <p>16.4.2 Potential Applications of T2D Predictive Models 387</p> <p>16.4.3 Barriers to the Adoption of T2D Predictive Models 387</p> <p>16.4.4 Addressing Practical Issues by Combining Multiple T2D Predictive Models 388</p> <p>16.4.5 The Combined Model Achieves High Prediction Performance with High Coverage 390</p> <p>16.5 Conclusions and Future Outlook 391</p> <p>Author Biography 392</p> <p>References 393</p> <p>Index 399</p>