New method fills MS lesions more realistically and naturally.
problem Lesions in MS brain scans bias morphometric analyses.
method Non-local partial convolutions (NLPC) integrating Unet-like network and non-local module.
result NLPC generates inpainted regions that appear more realistic and natural.
Estimates localized complexity of white-matter wiring using GANs.
problem Complexity of white-matter wiring, especially in ambiguous regions, confounds analysis.
method Bayesian estimate of heteroscedastic aleatoric uncertainty through image inpainting.
result Localized wiring complexity quantified, directly reflecting difficulty of lesion inpainting.
This project applies Mask R-CNN method to ISIC 2018 challenge tasks: lesion boundary segmentation (task1), lesion attributes detection (task 2), lesion diagnosis (task 3), a solution to the latter is using a trained model for task 1 and a simple voting procedure.
Automated detection of MS lesions improves to 67% with 7T MRI.
problem Accurate detection of small, scarce cortical lesions in MS patients.
method 3D U-Net with brain tissue segmentation, supervised training on 7T MRI.
result 67% lesion detection rate with 42% false positives.
Improved inter-scanner MS lesion segmentation through adversarial training.
problem Variability in MRI scanner or protocol differences affect automated lesion segmentation accuracy.
method Trained a CNN base model and a discriminator model adversarially on multi-scanner longitudinal data.
result Adversarial training improves inter-scanner consistency of lesion segmentations.
Deep neural networks map brain lesions to deficits for better brain function understanding.
problem Mapping the functional brain organization from pathological lesions.
method Deep generative neural network architectures, specifically variational convolutional volumetric auto-encoders.
result Our model outperforms established methods in lesion-deficit inference across various scenarios.
Paper tackles dermoscopy lesion boundary detection using deep learning.
problem Automatic detection of lesion boundaries in dermoscopy images.
method Adapted U-net convolutional neural network with skip connections.
result Demonstrates effectiveness of deep learning in lesion boundary segmentation.
Paper tackles rDR classification and lesion segmentation using self-supervised equivariant learning and attention-based MIL.
problem Classifying rDR and segmenting lesions from image-level labels.
method Integrates self-supervised equivariant attention mechanism (SEAM) with attention-based multi-instance learning (MIL).
result Achieved AU ROC of 0.958 on Eyepacs dataset, outperforming state-of-the-art.
CNN automates vitiligo lesion segmentation quickly and accurately.
problem Manual segmentation of vitiligo lesions is time-consuming and inconsistent.
method U-Net architecture with modified contracting path, followed by watershed algorithm refinement.
result CNN achieves 73.6% Jaccard Index, significantly outperforming state-of-the-art methods.
Semantic inpainting is the task of inferring missing pixels in an image given surrounding pixels and high level image semantics. Most semantic inpainting algorithms are deterministic: given an image with missing regions, a single inpainted image is generated. However, there are often several plausible inpaintings for a…
Theoretical justification for image inpainting using diffusion models.
problem Improving sample recovery in image inpainting without retraining.
method Analysis of RePaint algorithm and proposing RePaint+ to correct misalignment. result RePaint+ algorithm provably recovers the true sample with linear convergence. LcGAN generates synthetic CT images for hemorrhagic lesion segmentation.
problem Scarce training data for hemorrhagic lesion segmentation.
method Lesion conditional Generative Adversarial Network (LcGAN) for synthetic image generation.
result Segmentation improved by 12.8% with synthetic data augmentation.
Proposes deep quantile regression for uncertainty estimation in lesion detection.
problem Uncertainty quantification in lesion detection for critical applications.
method Quantile regression for aleatoric uncertainty, Variational AutoEncoder (VAE) with QR-VAE, binary quantile regression (BQR).
result Effective quantification of uncertainty in lesion detection and segmentation.
Generative model boosts bone lesion classification with augmented data.
problem Class imbalance and limited training data for bone lesion classification.
method Cycle-consistent GAN for image patch translation across bones.
result Generative augmentation improves classifier performance on held-out test set.
AI detects oral pre-cancerous lesions with high accuracy.
problem Manual screening of oral cavity cancer is expensive and lacks specialists.
method Deep convolutional neural networks (DCNNs) using transfer learning.
result DCNN models achieve high accuracy in distinguishing between benign and pre-cancerous tongue lesions.
Physics-informed semantic inpainting improves geostatistical modeling by incorporating indirect measurements.
problem Inferring heterogeneous geological fields from limited measurements and prior spatial statistics.
method Proposes a physics-informed semantic inpainting framework using WGAN-GP to incorporate both direct and indirect measurements.
result The method satisfies physical conservation laws and enhances inpainting performance compared to using only direct measurements.
In this work, we present a comparison of a shallow and a deep learning architecture for the automated segmentation of white matter lesions in MR images of multiple sclerosis patients. In particular, we train and test both methods on early stage disease patients, to verify their performance in challenging conditions, mo…
New method separates objects from images using deep neural networks trained to inpaint.
problem Fully self-supervised instance separation of occluded objects in images.
method Maximizes independence of two image regions given a fully self-supervised inpainting network.
result Method achieves similar segmentation performance to fully supervised methods on microscopy image datasets.
This paper proposes an innovative method for segmentation of skin lesions in dermoscopy images developed by the authors, based on fuzzy classification of pixels and histogram thresholding.
Paper presents an active learning method to reduce skin lesion analysis annotation cost.
problem High cost of manual annotation in skin lesion analysis.
method Decoupled sample selection criteria for informativeness and representativeness, combined with pixel space aggregation for data augmentation.
result 50% annotated data suffices for state-of-the-art performance, outperforming other methods.
GACELA fills long gaps in musical audio with a GAN and context conditioning.
problem Restoring long gaps in musical audio with varying complexity and duration.
method Generative adversarial network (GAN) with five parallel discriminators and context conditioning.
result Reduced artifacts in inpaintings from unacceptable to mildly disturbing.
This paper uses CNNs to automatically segment ischaemic stroke lesions from MRI sequences.
problem Automatically segmenting ischaemic stroke lesions from MRI sequences is challenging.
method Adversarial training of CNNs on multi-sequence MRI data.
result The method achieves high Dice scores for core and penumbra segmentation.
Computer-aided diagnosis systems for classification of different type of skin lesions have been an active field of research in recent decades. It has been shown that introducing lesions and their attributes masks into lesion classification pipeline can greatly improve the performance. In this paper, we propose a framew…
Efficient skin lesion analysis combines deep CNN and handcrafted features.
problem Skin lesion analysis for melanoma detection.
method Combines deep learning and handcrafted features, introduces hybrid features.
result 0.841 score on validation dataset using SVM classifier.
EnSF uses image inpainting to handle partial observations in data assimilation.
problem Data assimilation challenges with partial observations.
method EnSF integrates image inpainting with diffusion models to predict unobserved states.
result EnSF successfully tracks SQG dynamics with partial observations.
Unsupervised image inpainting models generate plausible reconstructions from incomplete data.
problem Inpainting without paired or unpaired training data.
method Conditional GAN with latent component dependency for generating image distributions.
result Model generates a distribution of plausible images from incomplete observations.
We address the problem of surface inpainting, which aims to fill in holes or missing regions on a Riemann surface based on its surface geometry. In practical situation, surfaces obtained from range scanners often have holes where the 3D models are incomplete. In order to analyze the 3D shapes effectively, restoring the…
Generative adversarial network improves audio inpainting for long gaps.
problem Generating missing audio content in long-range gaps using WGAN.
method Proposed WGAN architecture with short-range and long-range neighboring borders.
result The proposed model outperforms classical WGAN in reconstructing high-frequency content.
PersonalizedUS assesses breast cancer risk with local coverage guarantees.
problem Manual BI-RADS scoring leads to unnecessary biopsies and mental health burden.
method Conformal prediction for precise, personalized risk estimates.
result Local coverage guarantees with high sensitivity and specificity.
Extended RDS filtering for positions and orientations, improving crossing structure enhancement and inpainting.
problem Enhancing and inpainting images with crossing structures.
method Extended RDS filtering to M2 space, using gauge frames to mitigate issues. result RDS filtering outperforms existing techniques in denoising and inpainting images with crossing structures.
Generative adversarial networks have been successfully applied to inpainting in natural images. However, the current state-of-the-art models have not yet been widely adopted in the medical imaging domain. In this paper, we investigate the performance of three recently published deep learning based inpainting models: co…
Dual-edge spatial Jacobian image graph for interpretable diabetic retinopathy grading
problem Automated diabetic retinopathy grading from color fundus photographs
method Dual-edge spatial-Jacobian image graph
result 0.8076 accuracy, 0.8312 quadratic weighted kappa, 0.5915 macro-F1, 0.9330 adjacent-grade accuracy
Deep learning segments spinal metastases in MR images.
problem Challenges in accurately segmenting spinal metastases in MR images.
method Used a U-Net-like architecture trained on 40 clinical cases.
result Average Dice scores up to 77.6% and mean sensitivity rates up to 78.9%.
Paper introduces tractographic feature to predict stroke outcomes.
problem Predicting stroke outcomes using lesion volume alone is limited.
method Tractographic feature combining lesion and connectome data.
result Tractographic feature outperforms stroke volume in predicting mRS grades.
Model generates music to connect missing parts, leveraging latent space of VAE.
problem Music inpainting from missing or lost information.
method Deep learning model using VAE latent space and RNN to traverse latent space conditioned on past and future musical contexts.
result Model generates meaningful music inpaintings, connecting musical excerpts.
This paper explains the method used in the segmentation challenge (Task 1) in the International Skin Imaging Collaboration's (ISIC) Skin Lesion Analysis Towards Melanoma Detection challenge held in 2018. We have trained a U-Net network to perform the segmentation. The key elements for the training were first to adjust …
We develop a method for user-controllable semantic image inpainting: Given an arbitrary set of observed pixels, the unobserved pixels can be imputed in a user-controllable range of possibilities, each of which is semantically coherent and locally consistent with the observed pixels. We achieve this using a deep generat…
Deep learning methods, and in particular convolutional neural networks (CNNs), have led to an enormous breakthrough in a wide range of computer vision tasks, primarily by using large-scale annotated datasets. However, obtaining such datasets in the medical domain remains a challenge. In this paper, we present methods f…
High predictive performance and ease of use and interpretability are important requirements for the applicability of a computer-aided diagnosis (CAD) to human reading studies. We propose a CAD system specifically designed to be more comprehensible to the radiologist reviewing screening breast MRI studies. Multiparametr…
New method denoises and fills in missing image data without clean training data.
problem Denoising and inpainting images with noisy, incomplete data.
method Robust Hadamard Autoencoders trained on noisy, incomplete data.
result Autoencoders can perform denoising and inpainting simultaneously.
CNNs help diagnose diabetic retinopathy by localizing lesions.
problem Diabetic retinopathy diagnosis requires identifying lesions in fundus images.
method Post-attention technique (Grad-CAM) on deep learning models' penultimate layer.
result InceptionV3 model achieves best performance and localizes lesions better.
Natural image statistics exhibit hierarchical dependencies across multiple scales. Representing such prior knowledge in non-factorial latent tree models can boost performance of image denoising, inpainting, deconvolution or reconstruction substantially, beyond standard factorial "sparse" methodology. We derive a large …
A novel deep learning architecture (XmasNet) based on convolutional neural networks was developed for the classification of prostate cancer lesions, using the 3D multiparametric MRI data provided by the PROSTATEx challenge. End-to-end training was performed for XmasNet, with data augmentation done through 3D rotation a…
CosmoVAE uses deep learning to fill in missing parts of the cosmic microwave background map.
problem Missing observations in the CMB map due to thermal dust noise.
method Variational autoencoder (VAE) trained with a new loss function.
result CosmoVAE achieves state-of-the-art performance in CMB map inpainting.
KeyIn discovers keyframes to simplify video analysis.
problem Extracting essential dynamics from videos interleaved with predictable details.
method A hierarchical Keyframe-Inpainter (KeyIn) model learns keyframes and fills in the rest.
result KeyIn finds informative keyframes across various video dynamics.
Study improves breast lesion segmentation with limited in vivo data using simulated and natural images.
problem Challenges in automatic breast lesion segmentation due to limited annotated data.
method Pre-training a segmentation network on simulated and natural images, followed by fine-tuning with limited in vivo data.
result Fine-tuning improves dice score by 21% with as little as 19 in vivo images.
Method screens similar capsule endoscopic images, reducing doctor workload and improving accuracy.
problem Time-consuming and high error rate in manual inspection of large numbers of similar capsule endoscopic images.
method Structural similarity analysis of visually salient areas and hierarchical clustering.
result 76% reduction in similar images, 100% lesion recall, 18-minute average play time.
In this paper we present an end-to-end deep learning framework to turn images that show dynamic content, such as vehicles or pedestrians, into realistic static frames. This objective encounters two main challenges: detecting all the dynamic objects, and inpainting the static occluded background with plausible imagery. …