LEADER 11753nam 22005415 450 001 9910881100203321 005 20251225202101.0 010 $a3-031-67285-2 024 7 $a10.1007/978-3-031-67285-9 035 $a(MiAaPQ)EBC31605056 035 $a(Au-PeEL)EBL31605056 035 $a(CKB)34039643800041 035 $a(DE-He213)978-3-031-67285-9 035 $a(EXLCZ)9934039643800041 100 $a20240814d2024 u| 0 101 0 $aeng 135 $aurcnu|||||||| 181 $ctxt$2rdacontent 182 $cc$2rdamedia 183 $acr$2rdacarrier 200 10$aArtificial Intelligence in Healthcare $eFirst International Conference, AIiH 2024, Swansea, UK, September 4?6, 2024, Proceedings, Part II /$fedited by Xianghua Xie, Iain Styles, Gibin Powathil, Marco Ceccarelli 205 $a1st ed. 2024. 210 1$aCham :$cSpringer Nature Switzerland :$cImprint: Springer,$d2024. 215 $a1 online resource (353 pages) 225 1 $aLecture Notes in Computer Science,$x1611-3349 ;$v14976 311 08$a3-031-67284-4 320 $aIncludes bibliographical references and index. 327 $aIntro -- Preface -- Organization -- Contents - Part II -- Contents - Part I -- AI in Proactive Care and Intervention -- Cluster and Trajectory Analysis of Multiple Long-Term Conditions in Adults with Learning Disabilities -- 1 Introduction -- 2 Methods -- 2.1 Dataset -- 2.2 Dataset Preparation -- 2.3 Cluster Analysis -- 3 Results -- 3.1 Identifying Clusters of MLTC for Adults with Learning Difficulties -- 3.2 What Were the Baseline Patient Characteristics Across the Identified Clusters? -- 3.3 What Were the Most Common Combinations of Multiple LTCs and the Most Frequent Trajectories for the Most Dominant Clusters? -- 4 Conclusion -- References -- A Deep Learning Framework for Assessing the Risk of Transvenous Lead Extraction Procedures -- 1 Introduction -- 2 Automatic Object Detection -- 2.1 Selecting Geometric Features -- 2.2 The Detection of the Approximate Location of SVC -- 2.3 The Detection of Pacing Leads and Coils -- 2.4 The Detection of Coils -- 3 The Extraction of Geometric Features -- 3.1 The Determination of the Coil Position Related to the SVC -- 3.2 The Lead Angulations -- 3.3 Detecting the Number of Leads in the SVC -- 4 Feature Selection and Machine Learning Model -- 4.1 Feature Selection -- 4.2 Machine Learning Model for Risk Assessment -- 5 Conclusions -- References -- AI-Aided Medical Imaging and Analysis -- Assessing the Impact of Deep Learning Backbones for Mass Detection in Breast Imaging -- 1 Introduction -- 2 Related Work -- 3 Materials and Methods -- 3.1 Dataset -- 3.2 Deep Learning Models -- 3.3 Training -- 3.4 Metrics -- 4 Results -- 4.1 Effect of Backbone Pre-training -- 4.2 Effect of Backbone Fine-Tuning Inside the Detection Network -- 4.3 Backbone Comparison -- 4.4 Performance Relative to the Size of the Network -- 4.5 Comparison with Other Works on CBIS-DDSM -- 5 Conclusion -- References. 327 $aTransferable Variational Feedback Network for Vendor Generalization in Accelerated MRI -- 1 Introduction -- 2 Materials and Methods -- 2.1 Background and Problem Formulation -- 2.2 Base Architecture: Variational Feedback Network -- 2.3 Proposed Feature Transfer Learning Architecture -- 3 Experiments and Results -- 3.1 Datasets -- 3.2 Training Protocol -- 3.3 Vendor Transfer -- 3.4 Learning Without Forgetting in Accelerated MRI -- 3.5 Further Discussion -- 4 Conclusion -- References -- CVD_Net: Head and Neck Tumor Segmentation and Generalization in PET/CT Scans Across Data from Multiple Medical Centers -- 1 Introduction -- 2 Background -- 3 Method -- 3.1 CNN Encoder -- 3.2 Domain-Specific Batch Normalization (DSBN) -- 3.3 Transfomer Encoder -- 3.4 CNN Decoder -- 4 Experiments -- 4.1 Dataset -- 4.2 Training and Testing -- 4.3 Results -- 5 Conclusions -- References -- Applying Deep Learning Based Super-Resolution to Knee Imaging -- 1 Introduction -- 2 SR Models -- 2.1 SRCNN -- 2.2 ExSRCNN -- 2.3 RBSRCNN -- 3 Dataset and Experimental Setup -- 4 Experimental Results -- 5 Conclusions -- References -- FM-LiteLearn: A Lightweight Brain Tumor Classification Framework Integrating Image Fusion and Multi-teacher Distillation Strategies -- 1 Introduction -- 2 Related Work -- 3 Method -- 3.1 Image Fusion Techniques(F-DCGAN) -- 3.2 Improvement Based on ResNet18(T-ResNet18) -- 3.3 Multi-teacher Knowledge Distillation(MT-KD) -- 4 Experimental Setup -- 4.1 Datasets -- 4.2 Implementation Details -- 4.3 Evaluation Metrics -- 4.4 Experimental Results -- 5 Conclusion -- References -- Towards Improving Single-Cell Segmentation in Heterogeneous Configurations of Cardiomyocyte Networks -- 1 Introduction -- 2 Background -- 3 Methodology -- 3.1 The HL-1 Cardiac Cell Network Imaging Dataset -- 3.2 Data Preparation -- 3.3 Cell Annotation. 327 $a3.4 Segmentation Algorithms and Training -- 3.5 Evaluation Metrics -- 4 Validation Results -- 5 Cellular Network Analysis -- 6 Conclusion -- References -- Texture Feature Analysis for Classification of Early-Stage Prostate Cancer in MpMRI -- 1 Introduction -- 2 Datasets -- 3 Methods -- 4 Results -- 4.1 Best-Performing Machine Learning Classifiers -- 4.2 Feature Value Ranges and Correlations -- 4.3 Feature Impact on Model Output: Shapley Values -- 4.4 RF Classifiers Trained Without Feature Selection -- 5 Conclusions -- References -- Medical Signal and Image Processing -- DELRecon: Depth Electrode Reconstruction Toolbox for Stereo-EEG -- 1 Introduction -- 2 Method -- 2.1 Toolbox Installation and Requirements -- 2.2 Module 1: Image Processing -- 2.3 Module 2: Electrode Localization -- 2.4 Module 3: Individual Contact Labelling -- 2.5 Toolbox Outputs -- 2.6 Dataset Used for Validation -- 3 Results -- 3.1 MRI and CT Image Processing -- 3.2 Initial SEEG Electrode Clustering -- 3.3 Iterative Tracking of SEEG Electrodes -- 3.4 Contact Labelling -- 3.5 Evaluation -- 4 Discussion -- 5 Conclusion -- References -- Segmenting Breast Ultrasound Scans Using a Generative Adversarial Network Embedding U-Net -- 1 Introduction -- 2 Background and Design Motivations -- 3 Resources and Methods -- 3.1 Dataset -- 3.2 Training the Model -- 3.3 Experimental Setup -- 3.4 Results -- 4 Conclusion -- References -- Enhancing Predictive Accuracy in Embryo Implantation: The Bonna Algorithm and its Clinical Implications -- 1 Introduction -- 2 Methodology -- 2.1 Dataset Collection and Preparation -- 2.2 Data Processing and Flow -- 2.3 Model Architecture and Implementation. -- 3 Results -- 3.1 Model Performance Evaluation -- 3.2 Confidence and Predictive Accuracy -- 4 Discussion and Conclusion -- References. 327 $aBacterial Behaviour Analysis Through Image Segmentation Using Deep Learning Approaches -- 1 Introduction -- 2 Literature Review -- 3 Data Description and Terminologies -- 3.1 Data Description -- 3.2 Explored Attributes of Bacteria -- 4 Methodology -- 5 Result and Analysis -- 6 Conclusion -- References -- Assisted Living Technology -- Innovations in Mosquito Identification: Integrating Deep Learning with Citizen Science -- 1 Introduction -- 2 Similar Works -- 3 Method -- 3.1 Dataset Properties -- 3.2 Data Preprocessing -- 3.3 Model Architecture -- 3.4 Training and Evaluation -- 4 Results -- 4.1 Accuracy and Generalization -- 4.2 Loss Analysis -- 4.3 Evaluation Metrics -- 4.4 Robustness Across Scenarios -- 5 Comparison with Existing Literature -- 6 Discussion -- 6.1 Addressing Unbalanced Class Distribution -- 6.2 Generalization to Citizen Science Data -- 6.3 Future Directions -- 6.4 Implications for Public Health -- 7 Conclusion -- References -- Action Recognition for Privacy-Preserving Ambient Assisted Living -- 1 Introduction -- 2 Related Works -- 2.1 Real-Time Performance of Skeleton-Based Action Recognition -- 2.2 Data Augmentation Strategies -- 3 Method -- 3.1 TD-GDSCN -- 3.2 Data Augmentation Techniques -- 4 Experiments -- 4.1 DatasetsThe University Research Ethics Committee Approved the Dataset. Approval ID: EPS21036. -- 4.2 Implementation Details -- 4.3 Data Augmentation Techniques -- 4.4 Evaluation of Computational Efficiency -- 4.5 Comparison to State-of-the-Art Methods -- 5 Conclusion -- References -- Digital Twinning, Virtual Pathology and Oncology -- Weight Perturbations for Simulating Virtual Lesions in a Convolutional Neural Network -- 1 Introduction -- 1.1 Human Visual Recognition -- 1.2 Computational Models -- 1.3 CNNs as Computational Models -- 1.4 Aim and Hypothesis of the Study -- 1.5 Five Incremental Goals. 327 $a2 Experiment Materials and Methods -- 2.1 Matching Task -- 3 Experiment Results -- 3.1 Accuracy -- 3.2 Reaction Times -- 3.3 TMS Experiment Discussion -- 4 Modeling -- 4.1 Modeling Approach -- 4.2 Modeling Results -- 5 Discussion -- 5.1 Hypothesis Questions -- 5.2 Summary -- References -- Using GANs to Visualise Class-Specific Features in Digital Histopathology Images -- 1 Introduction -- 2 Literature Review -- 3 Methodology -- 3.1 Dataset -- 3.2 Training -- 4 Experimentation -- 4.1 Domain Representation - Comparison Between Models -- 4.2 Domain Representation - Class-Specific Representation of HPV Status -- 4.3 Linear Interpolation -- 5 Results -- 5.1 Domain Representation - Comparison Between Models -- 5.2 Domain Representation - Class Specific Representation of HPV Status -- 5.3 Linear Interpolation - Transition from HPV- to HPV+ -- 6 Discussion -- 7 Conclusion -- References -- Artificial Intelligence for Predicting Responses to Thyroid Cancer Treatment -- 1 Introduction -- 2 Methods -- 2.1 Dataset Information -- 2.2 Data Preprocessing -- 2.3 Experiments -- 2.4 Model Development and Evaluation -- 3 Results -- 3.1 Overall Performance -- 3.2 Performance Based on the Number of Classes -- 3.3 Performance Based on Features Used -- 4 Discussion -- 4.1 Principal Findings -- 4.2 Practical and Research Implications -- 4.3 Limitations -- 5 Conclusion -- References -- Patient Data, Privacy and Ethics -- ZMAM: A ZKP-Based Mutual Authentication Scheme for the IoMT -- 1 Introduction -- 2 Related Work -- 3 Background, Threat Model and Assumptions -- 3.1 Background: Zero-Knowledge Proof, Trusted Boot -- 3.2 Threat Model and Assumption -- 4 ZMAM Scheme Design -- 4.1 System Components -- 4.2 Security Goals -- 4.3 Registration Protocol (RP) -- 4.4 Normal Communication Protocol -- 4.5 Communication in Emergency Protocol -- 5 Security Analysis -- 5.1 Formal Analysis. 327 $a5.2 Informal Analysis. 330 $aThe two-volume set LNCS 14975 + 14976 constitutes the proceedings of the First International Conference on Artificial Intelligence in Healthcare, AIiH 2024, which took place in Swansea, UK, in September 2024. The 47 full papers included in the proceedings were carefully reviewed and selected from 70 submissions. They were organized in the following topical sections: Part I: Personalised Healthcare and Medicine; AI driven early diagnosis and prevention; AI driven robotics for healthcare; AI in mental health; Part II: AI in proactive care and intervention; AI-aided medical imaging and analysis; Medical signal and image processing; Assisted living technology; Digital twinning, virtual pathology and oncology; Patient data, privacy and ethics. 410 0$aLecture Notes in Computer Science,$x1611-3349 ;$v14976 606 $aArtificial intelligence 606 $aArtificial Intelligence 615 0$aArtificial intelligence. 615 14$aArtificial Intelligence. 676 $a610.28563 702 $aXie$b Xianghua 801 0$bMiAaPQ 801 1$bMiAaPQ 801 2$bMiAaPQ 906 $aBOOK 912 $a9910881100203321 996 $aArtificial intelligence in healthcare$92901849 997 $aUNINA