Computer-aided analysis of gastrointestinal videos / / Jorge Bernal, Aymeric Histace, editors

• Pubblicazione: Cham, Switzerland : , : Springer, , [2021]; ©2021
• Descrizione fisica: 1 online resource (192 pages)
• Disciplina: 612.3
• Soggetto topico: Gastrointestinal system - Growth; Computer vision
• Persona (resp. second.): BernalJorge; HistaceAymeric
• Nota di contenuto: Intro -- Foreword -- Acknowledgements -- Introduction -- Contents -- List of Figures -- List of Tables -- Part I Clinical Context -- 1 Clinical Context for Intelligent Systems in Colonoscopy -- 1.1 Risk Factors -- 1.2 Pathogenesis -- 1.3 Management -- 1.4 Current Limitations of White Light Endoscopy -- References -- 2 Clinical Context for Wireless Capsule Endoscopy Image Analysis -- 2.1 Introduction -- 2.2 WCE Technical Aspects -- 2.3 Challenges in WCE -- 2.4 Conclusions -- References -- Part II Technical Context -- 3 Technical Context for Intelligent Systems in Colonoscopy -- 3.1 State of the Art on Polyp Detection -- 3.1.1 Handcrafted Methods -- 3.1.2 Machine Learning Algorithms -- 3.2 State of the Art on Polyp Segmentation -- 3.3 Technical Challenges -- 3.3.1 Polyp Detection and Localization -- 3.3.2 Polyp Segmentation -- References -- 4 Technical Context for Wireless Capsule Endoscopy Image Analysis -- 4.1 WCE and Small Bowel -- 4.2 Vascular and Inflammatory Lesions in SB -- 4.3 Remaining Challenges -- 4.4 Conclusions -- References -- Part III Methodologies -- 5 Combination of Color-Based Segmentation, Markov Random Fields and Multilayer Perceptron -- 5.1 Motivation -- 5.2 Methodology -- 5.2.1 Pre-processing -- 5.2.2 Segmentation -- 5.3 Feature Extraction + Classification -- 5.4 Results -- References -- 6 Hand Crafted Method: ROI Selection and Texture Description -- 6.1 Motivation -- 6.2 ROI Selection Stage -- 6.2.1 The Convolutional Kernels -- 6.3 ROI Follow-Up Stage -- 6.4 ROI Description Stage -- 6.5 ROI Classification Stage -- 6.5.1 Fuzzy Trees -- 6.5.2 Fuzzy Forest -- 6.6 Results -- References -- 7 AECNN: Adversarial and Enhanced Convolutional Neural Networks -- 7.1 Introduction -- 7.2 Methodology -- 7.3 Generative Adversarial Networks -- References -- 8 Dilated ResFCN and SE-Unet for Polyp Segmentation -- 8.1 Motivation. ; 8.2 Introduction of the Base Structure -- 8.3 Methodology Explanation -- 8.3.1 Dilated ResFCN -- 8.3.2 SE-Unet -- 8.3.3 Training-Time Data Augmentation -- 8.3.4 Test-Time Data Augmentation -- 8.4 Example of Results -- References -- 9 Multi-encoder Decoder Network for Polyp Detection -- 9.1 Motivation -- 9.2 Y-Net for Polyp Detection -- 9.2.1 Model Learning and Implementation -- 9.3 Example of Performance -- References -- 10 Multi-resolution Multi-task Network and Polyp Tracking -- 10.1 Motivation -- 10.2 Introduction of the Base Structure -- 10.3 Multi-resolution Guidance and Class Activation Map Supervision -- 10.4 Bag of Losses -- 10.5 Polyp Tracking Algorithm -- 10.6 Data Normalization and Augmentation -- 10.7 Example of Results -- References -- 11 Region-Based Convolutional Neural Network for Polyp Detection and Segmentation -- 11.1 Introduction -- 11.2 ResNet 101 and Inception-ResNet-v2 -- 11.3 Polyp Detection -- 11.4 Polyp Segmentation -- 11.5 Training the Systems -- 11.5.1 Augmentations and Fine-Tuning -- 11.5.2 Objective Losses -- References -- 12 ResNet -- 12.1 General Motivation -- 12.2 Introduction to ResNet Architecture -- 12.3 Methodologies -- 12.3.1 RTC-ATC Group -- 12.3.2 Neuromation -- 12.3.3 Konica Minolta -- 12.4 Examples of Results (on the Training Sets) -- 12.4.1 RTC-ATC -- 12.4.2 Neuromation -- 12.4.3 Konica Minolta -- References -- 13 Multi-scale Ensemble of ResNet Variants -- 13.1 Motivation -- 13.2 Methods -- 13.2.1 Pre-processing -- 13.2.2 Class Split -- 13.2.3 Training Details -- 13.2.4 Ensembling -- 13.2.5 Test-Time Augmentation -- 13.2.6 Validation -- References -- 14 Convolutional LSTM -- 14.1 Introduction -- 14.2 Methods -- 14.3 Task-Specific Parameter Tuning -- 14.3.1 Polyp Detection -- 14.3.2 Polyp Localization -- 14.3.3 Polyp Segmentation -- 14.3.4 Polyp Segmentation HD -- References -- 15 TernausNet. ; 15.1 Model Architecture -- 15.2 Model Training -- 15.3 Postprocessing -- 15.4 Results -- References -- 16 Regression-Based Convolutional Neural Network with a Tracker -- 16.1 Motivation -- 16.2 Method Description in Details -- 16.2.1 Spatial Feature Learning -- 16.2.2 Temporal Information Integration -- 16.2.3 Experimental Setup -- References -- 17 Other Methodologies -- 17.1 GastroView Angiodysplasia Detection and Localization -- 17.1.1 Base Model -- 17.1.2 Detection -- 17.1.3 Localization -- References -- Part IV Experimental Setup -- 18 Polyp Detection in Colonoscopy Videos -- 18.1 CVC-VideoClinicDB Dataset -- 18.2 Performance Metrics -- 18.3 Validation Experiments -- 18.4 Participating Teams -- Reference -- 19 Polyp Segmentation in Colonoscopy Images -- 19.1 CVC-HDSegment, CVC-300 Y CVC-612 Datasets -- 19.2 Performance Metrics -- 19.3 Validation Experiments -- 19.4 Participating Teams -- References -- 20 Wireless Capsule Endoscopy Image Analysis -- 20.1 CAD-CAP Database -- 20.2 Performance Metrics -- 20.2.1 GIANA 2017 -- 20.2.2 GIANA 2018 -- 20.3 Participating Teams -- Part V Experimental Results and Analysis -- 21 Polyp Detection in Colonoscopy Videos -- 21.1 Polyp Detection and Localization -- 21.1.1 GIANA 2017 Challenge -- 21.1.2 GIANA 2018 Challenge -- 21.2 Evolution of Results -- 22 Polyp Segmentation in Colonoscopy Images -- 22.1 Polyp Segmentation SD -- 22.2 Polyp Segmentation HD -- 22.3 Evolution of Results -- 23 Wireless Capsule Endoscopy Image Analysis -- 23.1 Introduction -- 23.2 2017: Angiodysplasia Detection and Localization -- 23.3 2018: Multilabel Detection and Localization -- 24 Conclusions and Perspectives -- 24.1 Colonoscopy Image Analysis -- 24.2 WCE Image Analysis -- 24.3 Overall Conclusions.
• Titolo autorizzato: Computer-aided analysis of gastrointestinal videos
• ISBN: 3-030-64340-9
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 996464528803316
• Lo trovi qui: Univ. di Salerno