Computer vision Books

Book SynopsisThis book constitutes the refereed proceedings of the Third MICCAI Workshop on Domain Adaptation and Representation Transfer, DART 2021, and the First MICCAI Workshop on Affordable Healthcare and AI for Resource Diverse Global Health, FAIR 2021, held in conjunction with MICCAI 2021, in September/October 2021. The workshops were planned to take place in Strasbourg, France, but were held virtually due to the COVID-19 pandemic.DART 2021 accepted 13 papers from the 21 submissions received. The workshop aims at creating a discussion forum to compare, evaluate, and discuss methodological advancements and ideas that can improve the applicability of machine learning (ML)/deep learning (DL) approaches to clinical setting by making them robust and consistent across different domains. For FAIR 2021, 10 papers from 17 submissions were accepted for publication. They focus on Image-to-Image Translation particularly for low-dose or low-resolution settings; Model Compactness and Compression; Domain Adaptation and Transfer Learning; Active, Continual and Meta-Learning. Table of ContentsDomain Adaptation and Representation Transfer.- A Systematic Benchmarking Analysis of Transfer Learning for Medical Image Analysis.- Self-supervised Multi-scale Consistency for Weakly Supervised Segmentation Learning.- FDA: Feature Decomposition and Aggregation for Robust Airway Segmentation.- Adversarial Continual Learning for Multi-Domain Hippocampal Segmentation.- Self-Supervised Multimodal Generalized Zero Shot Learning For Gleason Grading.- Self-Supervised Learning of Inter-Label Geometric Relationships For Gleason Grade Segmentation.- Stop Throwing Away Discriminators! Re-using Adversaries for Test-Time Training.- Transductive image segmentation: Self-training and effect of uncertainty estimation.- Unsupervised Domain Adaptation with Semantic Consistency across Heterogeneous Modalities for MRI Prostate Lesion Segmentation.- Cohort Bias Adaptation in Federated Datasets for Lesion Segmentation.- Exploring Deep Registration Latent Spaces.- Learning from Partially Overlapping Labels: Image Segmentation under Annotation Shift.- Unsupervised Domain Adaption via Similarity-based Prototypes for Cross-Modality Segmentation.- A ordable AI and Healthcare.- Classification and Generation of Microscopy Images with Plasmodium Falciparum via Arti cial Neural Networks using Low Cost Settings.- Contrast and Resolution Improvement of POCUS Using Self-Consistent CycleGAN.- Low-Dose Dynamic CT Perfusion Denoising without Training Data.- Recurrent Brain Graph Mapper for Predicting Time-Dependent Brain Graph Evaluation Trajectory.- COVID-Net US: A Tailored, Highly Efficient, Self-Attention Deep Convolutional Neural Network Design for Detection of COVID-19Patient Cases from Point-of-care Ultrasound Imaging.- Inter-Domain Alignment for Predicting High-Resolution Brain Networks Using Teacher-Student Learning.- Sickle Cell Disease Severity Prediction from Percoll Gradient Images using Graph Convolutional Networks.- Continual Domain Incremental Learning for Chest X-ray Classification in Low-Resource Clinical Settings.- Deep learning based Automatic detection of adequately positioned mammograms.- Can non-specialists provide high quality Gold standard labels in challenging modalities.

Read more less

Book SynopsisAs the first book of a three-part series, this book is offered as a tribute to pioneers in vision, such as Béla Julesz, David Marr, King-Sun Fu, Ulf Grenander, and David Mumford. The authors hope to provide foundation and, perhaps more importantly, further inspiration for continued research in vision. This book covers David Marr's paradigm and various underlying statistical models for vision. The mathematical framework herein integrates three regimes of models (low-, mid-, and high-entropy regimes) and provides foundation for research in visual coding, recognition, and cognition. Concepts are first explained for understanding and then supported by findings in psychology and neuroscience, after which they are established by statistical models and associated learning and inference algorithms. A reader will gain a unified, cross-disciplinary view of research in vision and will accrue knowledge spanning from psychology to neuroscience to statistics. Table of ContentsPreface.- About the Authors.- 1 Introduction.- 2 Statistics of Natural Images.- 3 Textures.- 4 Textons.- 5 Gestalt Laws and Perceptual Organizations.- 6 Primal Sketch: Integrating Textures and Textons.- 7 2.1D Sketch and Layered Representation.- 8 2.5D Sketch and Depth Maps.- 9 Learning about information Projection.- 10 Informing Scaling and Regimes of Models.- 11 Deep Images and Models.- 12 A Tale of Three Families: Discriminative, Generative and Descriptive Models.- Bibliography

Read more less

Book SynopsisSelective Functional Connectivity between Ocular Dominance Columns in the Primary Visual Cortex.- ETSCL: An Evidence Theory-Based Supervised Contrastive Learning Framework for Multi-modal Glaucoma Grading.- VNR-AV: Structural Post-processing for Retinal Arteries and Veins Segmentation.- Wavelet Deep Learning Network for Objective Retinal Functional Estimation from Multimodal Retinal Imaging.- Inter-Frame Sclera Vessel Rotation Tracking for Toric Intraocular Lens Implantation Navigation.- Data Heterogeneity-aware Personalized Federated Learning for Diagnosis.- MM-UNet: A Mixed MLP Architecture for Improved Ophthalmic Image Segmentation.- Coral-CVDs: A consistent ordinal regression model for cardiovascular diseases grading.- Affordable Deep Learning for Diagnosing Inherited and Common Retinal Diseases via Color Fundus Photography.- Comparative Analysis of Data Augmentation for Retinal OCT Biomarker Segmentation.- Advanced Diabetic Retinopathy Classification: Integrating Pathological Indicators Segmentation and Morphological Feature Analysis.- Masked Image Modelling for Retinal OCT Understanding.- A Dual-Stream Network for Langerhans' Cells Segmentation in CCM Images.- Formula-Driven Data Augmentation and Partial Retinal Layer Copying for Retinal Layer Segmentation.- Enhancing Community Vision Screening: AI-Driven Retinal Photography for Early Disease Detection and Patient Trust.- Enhancing Large Foundation Models to Identify Fundus Diseases Based on Contrastive Enhanced Low-Rank Adaptation Prompt.

Read more less

Book Synopsis.- Synthetic Augmentation for Anatomical Landmark Localization using DDPMs..- AdaptDiff: Cross-Modality Domain Adaptation via Weak Conditional Semantic Diffusion for Retinal Vessel Segmentation..- Adapted nnU-Net: A Robust Baseline for Cross-Modality Synthesis and Medical Image Inpainting..- Beyond MR Image Harmonization: Resolution Matters Too..- Benchmarking Robustness of Endoscopic Depth Estimation with Synthetically Corrupted Data..- A dual-task mutual learning framework for predicting post-thrombectomy cerebral haemorrhage..- TSynD: Targeted Synthetic Data Generation for Enhanced Medical Image Classification..- Beyond Intensity Transforms: Medical Image Synthesis Under Large Deformation..- Sim2Real in endoscopy segmentation with a novel structure aware image translation..- Fireflies: Photorealistic Simulation and Optimization of Structured Light Endoscopy..- Exogenous Agent-Free Synthetic Post-Contrast Imaging with a Cascade of Deep Networks for enhancement Prediction after Tumor Resection. A Parametric-Map Oriented Approach..- OCT Scans Simulation Framework for Data Augmentation and Controlled Evaluation of Signal Processing Approaches..- Enhancing Quantitative Image Synthesis through Pretraining and Resolution Scaling for Bone Mineral Density Estimation from a Plain X-ray Image..- Latent Pollution Model: The Hidden Carbon Footprint in 3D Image Synthesis..- Synthesizing Scalable CFD-Enhanced Aortic 4D Flow MRI for Assessing Accuracy and Precision of Deep-Learning Image Reconstruction and Segmentation Tasks..- MedEdit: Counterfactual Diffusion-based Image Editing on Brain MRI..- Using MR physics for domain generalisation and super-resolution..- Single-scan mpMRI Calibration of Multi-Species Brain Tumor Dynamics with Mass Effect..- Annotated Biomedical Video Generation using Denoising Diffusion Probabilistic Models and Flow Fields.

Read more less

Book SynopsisA Novel Momentum-Based Deep Learning Techniques for Medical Image Classification and Segmentation.- Generalizable Lymph Node Metastasis Prediction in Pancreatic Cancer.- IRUM: An Image Representation and Unified Learning Method for Breast Cancer Diagnosis from Multi-view Ultrasound Images.- Classification, Regression and Segmentation directly from k-Space in Cardiac MRI.- DDSB: An Unsupervised and Training-free Method for Phase Detection in Echocardiography.- Mitral Regurgitation Recogniton based on Unsupervised Out-of-Distribution Detection with Residual Diffusion Amplification.- Deep Reinforcement Learning with Multiple Centerline-Guidance for Localization of Left Atrial Appendage Orifice from CT Images.- Lung-CADex: Fully automatic Zero-Shot Detection & Classification of Lung Nodules in Thoracic CT Images.- CIResDiff: A Clinically-Informed Residual Diffusion Model for Predicting Idiopathic Pulmonary Fibrosis Progression.- Vision Transformer Model for Automated End-to-End Radiographic Assessment of Joint Damage in Psoriatic Arthritis.- CorticalEvolve: Age-Conditioned Ordinary Differential Equation Model for Cortical Surface Reconstruction.- CSR-dMRI: Continuous Super-Resolution of Diffusion MRI with Anatomical Structure-assisted Implicit Neural Representation Learning.- Atherosclerotic plaque stability prediction from longitudinal ultrasound images.- Leveraging IHC Staining to Prompt HER2 Status Prediction from HE-Stained Histopathology Whole Slide Images.- VIMs: Virtual Immunohistochemistry Multiplex staining via Text-to-Stain Diffusion Trained on Uniplex Stains.- Structural-Connectivity-guided Functional Connectivity Representation for Multi-modal Brain Disease Classification.-Clinical Brain MRI Super-Resolution with 2D Slice-Wise Diffusion Model.- Low-to-high Frequency Progressive K-Space Learning for MRI Reconstruction.- LSST: Learned Single-Shot Trajectory and Reconstruction Network for MR Imaging.- 7T-like T1-weighted and TOF MRI synthesis from 3T MRI with Multi-contrast Complementary Deep Learning.- A Probabilistic Hadamard U-Net for MRI Bias Field Correction.- Structure-Preserving Diffusion Model for Unpaired Medical Image Translation.- Simultaneous Image Quality Improvement and Artefacts Correction in Accelerated MRI.- Full-TrSUN: A Full-Resolution Transformer UNet for high quality PET image synthesis.- TS-SR3: Time-strided Denoising Diffusion Probabilistic Model for MR Super-resolution.- PDM: A Plug-and-Play Perturbed Multi-path Diffusion Module for Simultaneous Medical Image Segmentation Improvement and Uncertainty Estimation.- DyNo: Dynamic Normalization based Test-Time Adaptation for 2D Medical Image Segmentation.-Accurate Delineation of Cerebrovascular Structures from TOF-MRA with Connectivity-Reinforced Deep Learning.- Learning Instance-Discriminative Pixel Embeddings Using Pixel Triplets.- Geo-UNet: A Geometrically Constrained Neural Framework for Clinical-Grade Lumen Segmentation in Intravascular Ultrasound.- Domain Influence in MRI Medical Image Segmentation: spatial versus k-space inputs.- Enhanced Small Liver Lesion Detection and Segmentation Using a Size-focused Multi-model Approach in CT Scans.- Generation and Segmentation of Simulated Total-Body PET Images.- Integrating Convolutional Neural Network and Transformer for Lumen Prediction along the Aorta Sections.- CSSD: Cross-Supervision and Self-Denoising for Hybrid-Supervised Hepatic Vessel Segmentation.- Calibrated Diverse Ensemble Entropy Minimization for Robust Test-Time Adaptation in Prostate Cancer Detection.- SpineStyle: Conceptualizing Style Transfer for Image-Guided Spine Surgery on Radiographs.- SGSR: Structure-Guided Multi-Contrast MRI Super-Resolution via Spatio-Frequency Co-Query Attention.- Knowledge Distillation based Dual-Branch Network for Whole Slide Image Analysis.- DHSampling: Diversity-based Hyperedge Sampling in GNN Learning with Application to Medical Imaging Classification.

Read more less

Book SynopsisRobust Box Prompt based SAM for Medical Image Segmentation.- Multi-task Learning Approach for Intracranial Hemorrhage Prognosis.- Mitigating False Predictions In Unreasonable Body Regions.- UniFed: A Universal Federation of a Mixture of Highly Heterogeneous Medical Image Classification Tasks.- Tackling domain generalization for out-of-distribution endoscopic imaging.- Benchmarking Dependence Measures to Prevent Shortcut Learning in Medical Imaging.- Selective Classifier Based Search Space Shrinking for Radiographs Retrieval.- Pseudo-Rendering for Resolution and Topology-Invariant Cortical Parcellation.- Partially Supervised Unpaired Multi-Modal Learning for Label-Efficient Medical Image Segmentation.- VIS-MAE: An Efficient Self-Supervised Learning Approach on Medical Image Segmentation and Classification.- Transformer-based Parameter Fitting of Models derived from Bloch-McConnell Equations for CEST MRI Analysis.- Probabilistic 3D Correspondence Prediction from Sparse Unsegmented Images.- StoDIP: Efficient 3D MRF image reconstruction with deep image priors and stochastic iterations.- Detection of Emerging Infectious Diseases in Lung CT based on Spatial Anomaly Patterns.- Data Alchemy: Mitigating Cross-Site Model Variability Through Test Time Data Calibration.- Noise-robust onformal prediction for medical image classification.- Identifying Critical Tokens for Accurate Predictions in Transformer-based Medical Imaging Models.-Resource-efficient Medical Image Analysis with Self-adapting Forward-Forward Networks.- SDF-Net: A Hybrid Detection Network for Mediastinal Lymph Node Detection on Contrast CT Images.- Arges: Spatio-Temporal Transformer for Ulcerative Colitis Severity Assessment in Endoscopy Videos.- Characterizing the Histology Spatial Intersections between Tumor-infiltrating Lymphocytes and Tumors for Survival Prediction of Cancers Via Graph Contrastive Learning.-Identifying Nonalcoholic Fatty Liver Disease and Adanced Liver Fibrosis from MRI in UK Biobank.- Explainable and Controllable Motion Curve Guided Cardiac Ultrasound Video Generation.

Read more less

Book SynopsisDCI-Net: Remote Sensing Image-based Object Detector.- CROSS-MODAL SHIP GROUNDING: TOWARDS LARGE MODEL FOR ENHANCED FEW-SHOT LEARNING.- STNet: Small Target Detection Network for IR Imagery.- FF-Yolo: A Feature-fusion Yolo model for Small Scale FODs detection in Airport Runways.- Weakly Aligned Multi-Spectral Pedestrian Detection via Cross-Modality Differential Enhancement and Multi-Scale Spatial Alignment.- CrackUDA: Incremental Unsupervised Domain Adaptation for Improved Crack Segmentation in Civil Structures.- DS MYOLO: A Reliable Object Detector Based on SSMs for Driving Scenarios.- Robust Single-Cam Surround View Object  Detection and Localization Using Memory Maps.- Exploring the Reliability of Foundation Model-Based Frontier Selection in Zero-Shot Object Goal Navigation.- Reliable Semantic Understanding for Real World Zero-shot Object Goal Navigation.- AllWeather-Net: Unified Image Enhancement for Autonomous Driving Under Adverse Weather and Low-Light Conditions.- Uni4DAL: A Unified Baseline for Multi-dataset 4D Auto-Labeling.- Dual-Attention Fusion Network with Edge and Content Guidance for Remote Sensing Images Segmentation.- Distortion Correction Sub-Network for Semantic Segmentation based on Deep Hough Transform.- MemoFlow: Modifying Explicit Motion of Inconsistency in Optical Flow.- Enhanced Brain Tumor Segmentation Using Preprocessing Techniques and 3D U-Net.- Joint Top-Down and Bottom-Up Frameworks for 3D Visual Grounding.- Anticipating Future Object Compositions without Forgetting.- SPK: Semantic and Positional Knowledge for Zero-shot Referring Expression Comprehension.- Can Language Improve Visual Features For Distinguishing Unseen Plant Diseases?.- Show Me the World in My Language: Establishing the First Baseline for Scene-Text to Scene-Text Translation.- iGrasp: An Interactive 2D-3D Framework for 6-DoF Grasp Detection.- Goal-Driven Transformer for Robot Behavior Learning from Play Data.- Adaptive Dynamic VSLAM: Refining Semantic-Geometric Fusion and Static Background Inpainting.- Hierarchical Visual Place Recognition with Semantic-guided Attention.- Dense Reconstruction and Localization in Scenes with Glass Surfaces Based on ORB-SLAM2.- Content-Aware Feature Upsampling for Voxel-based 3D Semantic Segmentation.- Enhancing 3D Referential Grounding by Learning Coarse Spatial Relationships.- PointGADM: Geometry Acquainted Deep Model for 3D Point Cloud Analysis.- CroMA: Cross-Modal Attention for Visual Question Answering in Robotic Surgery.

Read more less

Book SynopsisPixel Embedding for Fractional Interpolation in Video Coding.- Scene Text Image Super-Resolution with CLIP Prior Guidance.- A Coverless Steganography of Face Privacy Protection with Diffusion Models.- CLIP-AGIQA: Boosting the Performance of AI-Generated Image Quality Assessment with CLIP.- CoDeiT: Contrastive Data-efficient Transformers for Deepfake Detection.- A Lightweight and High-Fidelity Model for Generalized Audio-Driven 3D Talking Face Synthesis.- Illuminating the Dark: Unpaired Retinex and FFT-Based Low-Light Image Enhancement.- PSTNet: A Progressive Sparse Transformer Network for Image Deraining.- Frequency Modulated Deformable Transformer for Underwater Image Enhancement.- Probing Attention-Driven Normalizing Flow Network for Low-Light Image Enhancement.- A Novel Encoder-Decoder Network with Multi-domain Information Fusion for Video Deblurring.- Self-Distilled Dual-Network with Pixel Screening Loss for Blind Image Deblurring.- Complementary Dual-Branch Network for Space-Time Video Super-Resolution.- SINGLE-IMAGE DRIVEN 3D VIEWPOINT TRAINING DATA AUGMENTATION FOR EFFECTIVE LABEL RECOGNITION.- Lightweight Single Image Super-Resolution Network Integrating CNN and Transformer.- A Synthetic Benchmarking Pipeline to Compare Camera Calibration Algorithms.- Arbitrary Clothing Style Transfer Based on Attention Mechanism.- Approximate Cuboidization of an Orthogonal Polyhedron: A Combinatorial Approach.- Enhancing Multi-Exposure High Dynamic Range Imaging with Overlapped Codebook for Improved Representation Learning.- Deformable Multi-Scale Network for Snow Removal in Video.- Fast Orthogonal Matching Pursuit through Successive Regression.- MPGTSRN: Scene Text Image Super-Resolution Guided by Multiple Visual-Semantic Prompts.- Connecting the Dots: Isolated Trails of Detected Narrow Rivers in Multispectral Images.- Peel and Pool: The Path to Mandala Perfection.- MCANet: Multimodal Caption Aware Training-free Video Anomaly Detection via Large Language Model.- 2by2: Weakly-Supervised Learning for Global Action Segmentation.- MDFIDNet: Multi Domain Feature Integration Denoising Network.- Dynamic Resolution Guidance for Facial Expression Recognition.

Read more less

Book Synopsis.- Single-source Domain Generalization for Coronary Vessels Segmentation in X-ray Angiography. .- Constraint-Based Model in Multimodal Learning to Improve Ventricular Arrhythmia Prediction. .- Automated estimation of cardiac stroke volumes from computed tomography. .- Peridevice leaks following left atrial appendage occlusion - analysis with morphology descriptive centerlines and explainable graph attention network. .- Improved 3D Whole Heart Geometry from Sparse CMR Slices. .- CavityBASNet: Cavity-focused Biatrial Automatic Segmentation on LGE MRI with augmented input channel and left-right myocardium splitting. .- A novel MRI-based electrophysiological computational model of progressive doxorubicin-induced fibrosis in the left ventricle. .- Quantitative comparison of blood flow patterns from in silico simulations and 4D flow data before and after left atrial occlusion. .- Panoramic anatomical context in 3D intracardiac echocardiography (ICE) with 3D registration and geometry-based image fusion. .- Physics-Informed Neural Networks can accurately model cardiac electrophysiology in 3D geometries and fibrillatory conditions. .- Beyond the standards: Fully-Automated Aortic Annulus Segmentation on Contrast-free Magnetic Resonance Imaging using a Computational Aorta Unwrapping Method. .- Coronary Artery Calcium Scoring from Non-Contrast Cardiac CT Using Deep Learning With External Validation. .- Effective approach based on student-teacher self-supervised deep learning for Multi-class Bi-Atrial Segmentation Challenge. .- Sampling-Pattern-Agnostic MRI Reconstruction through Adaptive Consistency Enforcement with Diffusion Model. .- HyperCMR: Enhanced Multi-Contrast CMR Reconstruction with Eagle Loss. .- A Multi-Contrast Cardiac MRI Reconstruction Method Using an Advanced Unrolled Network Architecture. .- Implicit Neural Representations for Registration of Left Ventricle Myocardium During a Cardiac Cycle. .- Deep Multi-contrast Cardiac MRI Reconstruction via vSHARP with Auxil iary Refinement Network. .- Multi-Model Ensemble Approach for Accurate Bi-Atrial Segmentation in LGE-MRI of Atrial Fibrillation Patients. .- Two-Stage nnU-Net for Automatic Multi-class Bi-Atrial Segmentation from LGE-MRIs. .- An Ensemble of 3D Residual Encoder UNet Models for Solving Multi-Class Bi-Atrial Segmentation Challenge. .- Evaluating Convolution, Attention, and Mamba Based U-Net Models for Multi-Class Bi-Atrial Segmentation from LGE-MRI. .- On the Foundation Model for Cardiac MRI Reconstruction. .- Multi-Loss 3D Segmentation for Enhanced Bi-Atrial Segmentation. .- Classification of Mitral Regurgitation from Cardiac Cine MRI using Clinically-Interpretable Morphological Features. .- Gaussian Process Emulators for Few-Shot Segmentation in Cardiac MRI. .- Global Control for Local SO(3)-Equivariant Scale-Invariant Vessel Segmentation. .- A self-distillation bi-atrial segmentation network for Cardiac MRI. .- Adaptive Unrolling Applied to the CMRxRecon2024 Callenge. .- Reducing the number of leads for ECG Imaging with Graph Neural Networks and meaningful latent space. .- Rotor Core Projection Ablation (RCPA): Novel Computational Approach to Catheter Ablation Therapy for Atrial Fibrillation. .- Automated pipeline for regional epicardial adipose tissue distribution analysis in the left atrium. .- Low-Rank Conjugate Gradient-Net for Accelerated Cardiac MR Imaging. .- SBAW-Net: Segmentation of Bi-Atria and Wall Network - Offering Valuable Insights into Challenge Data. .- ResNet-based Convolutional Framework for Segmenting Left Atrial Scars and Cavities. .- EAT-Mamba: Epicardial Adipose Tissue Segmentation from Multi-modal Dixon MRI. .- Neural Fields for Continuous Periodic Motion Estimation in 4D Cardiovascular Imaging. .- Exploring CNN and Transformer Architectures for Multi-class Bi-Atrial Segmentation from Late Gadolinium-Enhanced MRI. .- EigenBoundaries for the temporally regularized segmentation of echocardiographic images. .- Dynamic Cardiac MRI Reconstruction via Separate Optimization of K-space and Hybrid-domian Spatial-temporal Feature Fusion. .- an Interpretable Learning of Risk Explain Ventricular Arrhythmia Mechanism. .- 3D Left Ventricular Reconstruction from 2D Echocardiograms for Reliable Volume Estimation. .- Comparing Left Atrial Spontaneous Echo Contrast Intensity with Gaussian Process Emulator Predictions. .- UPCMR: A Universal Prompt-guided Model for Random Sampling Cardiac MRI Reconstruction. .- An All-in-one Approach for Accelerated Cardiac MRI Reconstruction. .- Improving the Scan-rescan Precision of AI-based CMR Biomarker Estimation.

Read more less

Book SynopsisIntroduction.- Background.- Research objectives and method.- Datasets.- Comparison of the reviewed methods.- Conclusion.

Read more less

Book SynopsisPart 1: Pre-training and Datasets.- Chapter 1: LAION-5B: A Massive Open Image-Text Dataset.- Chapter 2: Efficient Training of Large-Scale Vision-Language Models.- Chapter 3: Scaling Laws for Contrastive Language-Image Learning.- Chapter 4: Scaling Up Vision-Language Models for Generic Tasks.- Chapter 5: Searching for Next-Gen Multimodal Datasets.- Part 2: Prompting and Generalization.- Chapter 6: Soft Prompt Learning for Vision-Language Models.- Chapter 7: Unified Prompting for Vision and Language.- Chapter 8: Zero-Shot Image Classification with Custom Prompts.- Chapter 9: Enhancing Vision-Language Models with Feature Adapters.- Chapter 10: Automatic Optimization of Prompting Architectures.- Chapter 11: Open-Vocabulary Calibration for VL Models.- Part 3: Applications.- Chapter 12: Open-Vocabulary DETR with Conditional Matching.- Chapter 13: Extracting Dense Labels from CLIP.- Chapter 14: PointCLIP: Understanding Point Clouds with VL.- Chapter 15: Diffusion-Based Relation Inversion from Images.- Chapter 16: Text-to-Video Generation.- Chapter 17: Text-Driven Human Motion Generation.- Chapter 18: Zero-Shot Text-Driven 3D Avatar Generation.- Chapter 19: Zero-Shot Text-Driven HDR Panorama Generation.

Read more less

Book SynopsisComputer-aided diagnosis/surgery: Concepts from Neurons: Building Interpretable Medical Image Diagnostic Models by Dissecting Opaque Neural Networks.- BioSonix: Can Physics-based Sonification Perceptualize Tissue Deformations from Tool Interactions? Brain: Explainable Deep Model for Understanding Neuropathological Events Through Neural Symbolic Regression.- A Multi-Layer Neural Transport Model for Characterizing Pathology Propagation in Neurodegenerative Diseases.- Enhancing Alzheimer's Diagnosis: Leveraging Anatomical Landmarks in Graph Convolutional Neural Networks on Tetrahedral Meshes.- Hierarchical Variable Importance with Statistical Control for Medical Data-Based Prediction.- Disentangle disease-relevant patterns from irrelevant patterns in fMRI analysis using equivariant and contrastive learning. Diffusion models: Continuous Diffusion Model for Self-supervised Denoising and Super-resolution on Fluorescence Microscopy Images.- Self-Supervised Denoising of Diffusion MRI Data with Efficient Collaborative Diffusion Model.- MAD-AD: Masked Diffusion for Unsupervised Brain Anomaly Detection. Self-supervised learning: Taming Masked Image Modeling for Chest X-ray Diagnosis by Incorporating Clinical Visual Priors.- Diffusion MAE: Paving the Way for Representation Learning of Diffusion MRI.- Resolving quantitative MRI model degeneracy in self-supervised machine learning. Vision-language models: Knowledge-enhanced Hyperbolic Language-Image Pretraining for Zero-shot Learning.- Structure Observation Driven Image-Text Contrastive Learning for Computed Tomography Report Generation.- Hierarchical CLIPs for Fine-grained Anatomical Lesion Localization from Whole-body PET/CT Images.- Multi-View and Multi-Scale Alignment for Contrastive Language-Image Pre-training in Mammography.- Interpretable Few-Shot Retinal Disease Diagnosis with Concept-Guided Prompting of Vision-Language Models.- Full Conformal Adaptation of Medical Vision-Language Models.- A Reality Check of Vision-Language Pre-training in Radiology: Have We Progressed Using Text? Shape analysis: ToothForge: Automatic Dental Shape Generation using Synchronized Spectral Embeddings.- LEDA: Log-Euclidean Diffeomorphism Autoencoder for Efficient Statistical Analysis of Diffeomorphisms.- CoRLD: Contrastive Representation Learning of Deformable Shapes in Images. Time-series image analysis: 4DRGS: 4D Radiative Gaussian Splatting for Efficient 3D Vessel Reconstruction from Sparse-View Dynamic DSA Images.- Brightness-Invariant Tracking Estimation in Tagged MRI.- SafeTriage: Facial Video De-identification for Privacy-Preserving Stroke Triage.

Read more less

Book SynopsisClassification/Detection: SpectMamba: Integrating Frequency and State Space Models for Enhanced Medical Image Detection.- Hierarchical Neural Cellular Automata for Lightweight Microscopy Image Classification.- PathTTT: Test-Time Training with Meta-Auxiliary Learning for Pathology Image Classification. Registration: Bi-invariant Geodesic Regression with Data from the Osteoarthritis Initiative.- GSSD: A Self-Distillation Paradigm with Gradient Surgery for End-to-End Deformable Image Registration.- Medical Image Registration Meets Vision Foundation Model: Prototype Learning and Contour Awareness.- Vascular-topology-aware Deep Structure Matching for 2D DSA and 3D CTA Rigid Registration.- Unsupervised Deformable Image Registration with Structural Nonparametric Smoothing. Reconstruction: Unsupervised Accelerated MRI Reconstruction via Ground-Truth-Free Flow Matching.- Optimization of acquisition schemes towards a better estimation of microstructure parameters in multidimensional diffusion MRI.- Bilinear Projector: Mitigating Discretization Artifacts in Model Based Iterative Reconstruction for X-ray CT.- Subspace Implicit Neural Representations for Real-Time Cardiac Cine MR Imaging. Image synthesis: 3D Shape-to-Image Brownian Bridge Diffusion for Brain MRI Synthesis from Cortical Surfaces.- Cascaded Diffusion Model and Segment Anything Model for Medical Image Synthesis via Uncertainty-Guided Prompt Generation.- DIReCT: Domain-Informed Rectified Flow for Controllable Brain MRI to PET Translation.- IGG: Image Generation Informed by Geodesic Dynamics in Deformation Spaces. Image enhancement: Cycle-consistent zero-shot through-plane super-resolution for anisotropic head MRI.- Bayesian Learning with Stochastic Perturbations and Langevin Expectation Maximization for Unsupervised DNN Image Quality Enhancement. Segmentation: MC-NuSeg: Multi-Contour Aware Nuclei Instance Segmentation with Segment Anything Model.- Pitfalls of topology-aware image segmentation.- GeoT: Geometry-guided Instance-dependent  Transition Matrix for Semi-supervised Tooth Point Cloud Segmentation.- RemInD: Remembering Anatomical Variations for Interpretable Domain Adaptive Medical Image Segmentation.- Dynamic Allocation Hypernetwork with Adaptive Model Recalibration for Federated Continual Learning.- SkeIite: Compact Neural Networks for Efficient Iterative Skeletonization.- VerSe: Integrating Multiple Queries as Prompts for Versatile Cardiac MRI Segmentation.

Read more less

Book Synopsis.- Frontiers in Computational Pathology..- Transductive Survival Ranking for Pan-cancer Automatic Risk Stratification using Whole Slide Images..- Benchmarking Histopathology Foundation Models in a Multi-center Dataset for Skin Cancer Subtyping..- MitoNet: Efficient Ki-67 Detection in H&E-Stained Images..- ASTER: Automated Segmentation of Endometrial Histology Images for Reproductive Health Assessment..- Leveraging Pathology Foundation Models for Panoptic Segmentation of Melanoma in H&E Images..- SMatt-DINO: Spatially Aware Masked Attention Network for High Resolution Brain Image Classification..- Persistent Homology and Gabor Features Reveal Inconsistencies Between Widely Used Colorectal Cancer Training and Testing Datasets..- SWIFT-Reg: Slide-Wide Intelligent Feature-based Tissue Registration..- Learnable Moran’s Index for Modeling Spatial Autocorrelation in Whole Slide Images to Predict Breast Cancer Outcomes..- Image Synthesis and Generative Artificial Intelligence..- Augmenting Chest X-ray Datasets with Non-Expert Annotations..- Leveraging Synthetic Data for Whole-Body Segmentation in X-ray Images..- Transform(AI)ng Radiology with CheXSBT: Integrating Dual-Attention Swin Transformer with BERT for Seamless Chest X-Ray Report Generation..- Cardiac Ultrasound Video Generation Using a Diffusion Model with Temporal Transformer..- KCLVA: Knowledge-enhanced Contrastive Learning and View-specific Attention for Chest X-ray Report Generation..- BlastDiffusion: A Latent Diffusion Model for Generating Synthetic Embryo Images to Address Data Scarcity in In Vitro Fertilization..- MediAug: Exploring Visual Augmentation in Medical Imaging..- On the Robustness of Medical Vision-Language Models: Are they Truly Generalizable?..- DiNO-Diffusion: Scaling Medical Diffusion Models via Self-Supervised Pre-Training..- Knowledge-Driven Hypothesis Generation for Burn Diagnosis from Ultrasound with Vision-Language Model..- Multimodal Federated Learning With Missing Modalities through Feature Imputation Network..- Parameter-Efficient Multimodal Adaptation for Certified Robustness of Medical Vision-Language Models.

Read more less

Book Synopsis.- Image-guided Diagnosis..- FD-SSD: Semi-Supervised Detection of Bone Fenestration and Dehiscence in Intraoral Images..- Interpretable Prediction of Lymph Node Metastasis in Rectal Cancer MRI Using Variational Autoencoders..- Self-Guided SwinTransformer Improves Breast Cancer Detection Through Iterative Attention-Based Zooming..- Can AI Be Faster, Accurate, and Explainable? SpikeNet Makes It Happen..- A Novel Feature-Prioritized Loss Function for Enhanced Pneumonia Segmentation in Chest X-rays..- Bridging Accuracy and Explainability: A SHAP-Enhanced CNN for Skin Cancer Diagnosis..- Multi-Scale WSI Analysis: A Cascade Framework for Efficient Breast Cancer Metastasis Detection..- Learning to Harmonize Cross-vendor X-ray Images by Non-linear Image Dynamics Correction..- Modified CBAM: Sub-Block Pooling for Improved Channel and Spatial Attention..- WSI-AL: A Novel Active Learning Framework for Whole Slide Image Selection..- A Deep-learning Approach for Diagnosing and Grading Ankylosing Spondylitis Sacroiliitis by X-ray Images..- Towards Breast Tumor Aggressiveness Classification in Digital Mammograms Using Boundary-Aware Segmentation and Feature Analysis..- Image-guided Intervention..- Joint Dento-Facial Shape Model..- Out-of-Distribution Detection in Gastrointestinal Vision by Estimating Nearest Centroid Distance Deficit..- Deep Learning-Driven Pipeline for Automated Wound Measurement of Chronic Wounds..- Midline-constrained Loss in the Anatomical Landmark Segmentation of 3D Liver Models..- DepthClassNet: A Multitask Framework for Monocular Depth Estimation and Texture Classification in Endoscopic Imaging..- Assessing the Generalization Performance of SAM for Ureteroscopy Scene Segmentation and Understanding..- Modelling Uncertainty in Graph Convolutional Networks for Edge Detection in Mammograms..- Classification of Gastroscopy Images Under extreme Class Imbalance: A Deep Learning Pipeline..- Temporally Consistent Smoke Removal from Endoscopic Video Images..- Toward Patient-specific Partial Point Cloud to Surface Completion for Pre- to Intra-operative Registration in Image-guided Liver Interventions..- EfficientDet with Knowledge Distillation and Instance Whitening for Real-time and Generalisable Polyp Detection.

Read more less

Book Synopsis.- Medical Image Segmentation..- TransE2UNet: Edge Guided TransEfficientUNET for Generalized Colon Polyp Segmentation from Endoscopy Images..- CA-Seg: An Attribute-based Medical Image Segmentation Framework for Unified Out-of-distributed Medical Image Segmentation..- TotalSegmentator 2D: A Tool for Rapid Anatomical Structure Analysis..- Promptable Cancer Segmentation Using Minimal Expert-curated Data..- SPARS: Self-Play Adversarial Reinforcement Learning for Segmentation of Liver Tumours..- Semantic Segmentation with Spreading Scribbles..- A Hybrid Transformer-Graph Model for Multi-Class Lymph Node Segmentation in Histopathology..- Exploring Context-Switching in Medical Image Retrieval Using Segmentation Models..- Segmentation in Histopathology Utilising Simulated Masked Patches..- A Feature-Driven Acquisition Strategy Using Scale-Invariant Descriptors for Deep Active Learning in Preclinical CT Segmentation..- Quantifying Inter-Annotator Agreement and Generalist Model Limitations in Imaging Mass Cytometry Single Cell Segmentation..- Subcortical Masks Generation in CT Images via Ensemble-Based Cross-Domain Label Transfer..- DRASU-Net: Dual-backbone and Residual Atrous Squeeze module-aided U-Net Model for Polyp Segmentation..- PolypDINO: Adapting DINOv2 for Domain Generalized Polyp Segmentation..- Intraoperative Segmentation Through Deep Learning and Mask Post-processing in Laparoscopic Liver Surgery..- Retinal and Vascular Image Analysis..- Hessian-based Deep Retinal Vessel Segmentation with Extremely Few Annotations..- Diffusion with Adversarial Fine-Tuning for Improving Rare Retinal Disease Diagnosis..- Deep Learning for Cardiovascular Risk Assessment: Proxy Features from Carotid Sonography as Predictors of Arterial Damage..- Enhanced Coronary Artery Segmentation in CTCA Using Bridging Centreline Integration..- QD-RetNet: Efficient Retinal Disease Classification via Quantized Knowledge Distillation..- Exploring the Effectiveness of Deep Features from Domain-Specific Foundation Models in Retinal Image Synthesis..- GenVOG: A Diffusion Probabilistic Framework for Patient-Independent Pose-Guided Nystagmus Video-Oculography (VOG) Generation..- Structurally Different Neural Network Blocks for the Segmentation of Atrial and Aortic Perivascular Adipose Tissue in Multi-centre CT Angiography Scans.

Read more less

Book Synopsis.- MedLoRD: A Medical Low-Resource Diffusion Model for High-Resolution 3D CT Image Synthesis..- GLFC: Unified Global-Local Feature and Contrast Learning with Mamba-EnhancedUNet for Synthetic CT Generation from CBCT..- 2D to 3D MR Image Super-Resolution using Cross-Contrast Guidance..- 3D Super-Resolution for Enhancing Compression Fracture Detection in Thick Slice CT: Diffusion Models vs GANs..- From Tissue-Mimicking Phantoms to Physics-Based Scans: Synthetic OCT forFew-Shot Foundation Model Training..- Multi-modal Brain MRI Synthesis with nnU-Net: Exploring Segmentation Per formance and Cross-Modality Relationships..- Unified 3D MRI Representations via Sequence-Invariant Contrastive Learning..- From Lines to Shapes: Geometric-Constrained Segmentation of X-Ray Collima tors via Hough Transform..- VIOLET: A framework for combined Volumetric Image registration via Opti mization and Learning for Efficient image Translation..- Generation of Controllable and Photorealistic Synthetic Cataract Surgery Im ages: Blending 3D Models and Real-World Data..- Unsupervised MRI Harmonization via Parameter Prediction and Super-ResolvedMPMs..- Learning Mechanistic Subtypes of Neurodegeneration with a Physics-InformedVariational Autoencoder Mixture Model..- FastDTI: A 3D Scale-arbitrary Super-resolution Autoencoder Residual DenseNetwork for DTI..- Lesion-Aware CT-to-MRI Synthesis using a Mask-Informed Diffusion with Adaptive Weighted Loss (MIDAS)..-Conditional Iterative α-(de)Blending Model for CBCT-tosCT Synthesis: Towards a Deterministic and Simple Process..- Synthesizing Accurate and Realistic T1-weighted Contrast-Enhanced MR Images using Posterior-Mean Rectified Flow..- Clustering-based Stain Augmentation: Templates for Periodic Acid-Schiff BiopsyImages.

Read more less

Book Synopsis.- LTCXNet: Tackling Long-Tailed Multi-Label Classification and RacialBias in Chest X-Ray Analysis..- Fairness and Robustness of CLIP-Based Models for Chest X-rays..- ShortCXR: Benchmarking Self-Supervised Learning Methods forShortcut Mitigation in Chest X-Ray Interpretation..- How Fair Are Foundation Models? Exploring the Role of CovariateBias in Histopathology..- The Cervix in Context: Bias Assessment in Preterm Birth Prediction..- Identifying Gender-Specific Visual Bias Signals in Skin Lesion Classification..- Fairness-Aware Data Augmentation for Cardiac MRI usingText-Conditioned Diffusion Models..- Exploring the interplay of label bias with subgroup size and separability:A case study in mammographic density classification..- Does a Rising Tide Lift All Boats? Bias Mitigation for AI-based CMRSegmentation..- MIMM-X: Disentangeling Spurious Correlations for Medical ImageAnalysis..- Predicting Patient Self-reported Race From Skin Histological Imageswith Deep Learning..- Robustness and sex differences in skin cancer detection: logisticregression vs CNNs..- Sex-based Bias Inherent in the Dice Similarity Coefficient: A ModelIndependent Analysis for Multiple Anatomical Structures..- The Impact of Skin Tone Label Granularity on the Performance andFairness of AI Based Dermatology Image Classification Models..- Causal Representation Learning with Observational Grouping for CXRClassification..- Invisible Attributes, Visible Biases: Exploring Demographic Shortcutsin MRI-based Alzheimer’s Disease Classification..- Fair Dermatological Disease Diagnosis through Auto-weightedFederated Learning and Performance-aware Personalization..- Assessing Annotator and Clinician Biases in an Open-Source-BasedTool Used to Generate Head CT Segmentations for Deep LearningTraining..- meval: A Statistical Toolbox for Fine-Grained Model Performance Analysis..- Revisiting the Evaluation Bias Introduced by Frame SamplingStrategies in Surgical Video Segmentation Using SAM2..- Disentanglement and Assessment of Shortcuts in OphthalmologicalRetinal Imaging Exams.

Read more less

Book Synopsis

Read more less