Search

Semantic segmentation has been widely investigated in the community, in which the state of the art techniques are based on supervised models. Those models have reported unprecedented performance at the cost of requiring a large set of high quality segmentation masks. To obtain such annotations is highly expensive and time consuming, in particular, in semantic segmentation where pixel-level annotations are required. In this work, we address this problem by proposing a holistic solution framed as a three-stage self-training framework for semi-supervised semantic segmentation. The key idea of our technique is the extraction of the pseudo-masks statistical information to decrease uncertainty in the predicted probability whilst enforcing segmentation consistency in a multi-task fashion. We achieve this through a three-stage solution. Firstly, we train a segmentation network to produce rough pseudo-masks which predicted probability is highly uncertain. Secondly, we then decrease the uncertainty of the pseudo-masks using a multi-task model that enforces consistency whilst exploiting the rich statistical information of the data. We compare our approach with existing methods for semi-supervised semantic segmentation and demonstrate its state-of-the-art performance with extensive experiments., Accepted for publication by IEEE Transactions on Image Processing

BACKGROUND: Advances in computer processing and improvements in data availability have led to the development of machine learning (ML) techniques for mammographic imaging. PURPOSE: To evaluate the reported performance of stand-alone ML applications for screening mammography workflow. MATERIALS AND METHODS: Ovid Embase, Ovid Medline, Cochrane Central Register of Controlled Trials, Scopus, and Web of Science literature databases were searched for relevant studies published from January 2012 to September 2020. The study was registered with the PROSPERO International Prospective Register of Systematic Reviews (protocol no. CRD42019156016). Stand-alone technology was defined as a ML algorithm that can be used independently of a human reader. Studies were quality assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and the Prediction Model Risk of Bias Assessment Tool, and reporting was evaluated using the Checklist for Artificial Intelligence in Medical Imaging. A primary meta-analysis included the top-performing algorithm and corresponding reader performance from which pooled summary estimates for the area under the receiver operating characteristic curve (AUC) were calculated using a bivariate model. RESULTS: Fourteen articles were included, which detailed 15 studies for stand-alone detection (n = 8) and triage (n = 7). Triage studies reported that 17%–91% of normal mammograms identified could be read by adapted screening, while “missing” an estimated 0%–7% of cancers. In total, an estimated 185 252 cases from three countries with more than 39 readers were included in the primary meta-analysis. The pooled sensitivity, specificity, and AUC was 75.4% (95% CI: 65.6, 83.2; P = .11), 90.6% (95% CI: 82.9, 95.0; P = .40), and 0.89 (95% CI: 0.84, 0.98), respectively, for algorithms, and 73.0% (95% CI: 60.7, 82.6), 88.6% (95% CI: 72.4, 95.8), and 0.85 (95% CI: 0.78, 0.97), respectively, for readers. CONCLUSION: Machine learning (ML) algorithms that demonstrate a stand-alone application in mammographic screening workflows achieve or even exceed human reader detection performance and improve efficiency. However, this evidence is from a small number of retrospective studies. Therefore, further rigorous independent external prospective testing of ML algorithms to assess performance at preassigned thresholds is required to support these claims. ©RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Whitman and Moseley in this issue.

The task of classifying mammograms is very challenging because the lesion is usually small in the high resolution image. The current state-of-the-art approaches for medical image classification rely on using the de-facto method for ConvNets - fine-tuning. However, there are fundamental differences between natural images and medical images, which based on existing evidence from the literature, limits the overall performance gain when designed with algorithmic approaches. In this paper, we propose to go beyond fine-tuning by introducing a novel framework called MorphHR, in which we highlight a new transfer learning scheme. The idea behind the proposed framework is to integrate function-preserving transformations, for any continuous non-linear activation neurons, to internally regularise the network for improving mammograms classification. The proposed solution offers two major advantages over the existing techniques. Firstly and unlike fine-tuning, the proposed approach allows for modifying not only the last few layers but also several of the first ones on a deep ConvNet. By doing this, we can design the network front to be suitable for learning domain specific features. Secondly, the proposed scheme is scalable to hardware. Therefore, one can fit high resolution images on standard GPU memory. We show that by using high resolution images, one prevents losing relevant information. We demonstrate, through numerical and visual experiments, that the proposed approach yields to a significant improvement in the classification performance over state-of-the-art techniques, and is indeed on a par with radiology experts. Moreover and for generalisation purposes, we show the effectiveness of the proposed learning scheme on another large dataset, the ChestX-ray14, surpassing current state-of-the-art techniques., 10 pages, 5 figures

International audience; Can one learn to diagnose COVID-19 under extreme minimal supervision? Since the outbreak of the novel COVID-19 there has been a rush for developing Artificial Intelligence techniques for expert-level disease identification on Chest X-ray data. In particular, the use of deep supervised learning has become the go-to paradigm. However, the performance of such models is heavily dependent on the availability of a large and representative labelled dataset. The creation of which is a heavily expensive and time consuming task, and especially imposes a great challenge for a novel disease. Semi-supervised learning has shown the ability to match the incredible performance of supervised models whilst requiring a small fraction of the labelled examples. This makes the semi-supervised paradigm an attractive option for identifying COVID-19. In this work, we introduce a graph based deep semi-supervised framework for classifying COVID-19 from chest X-rays. Our framework introduces an optimisation model for graph diffusion that reinforces the natural relation among the tiny labelled set and the vast unlabelled data. We then connect the diffusion prediction output as pseudo-labels that are used in an iterative scheme in a deep net. We demonstrate, through our experiments, that our model is able to outperform the current leading supervised model with a tiny fraction of the labelled examples. Finally, we provide attention maps to accommodate the radiologist's mental model, better fitting their perceptual and cognitive abilities. These visualisation aims to assist the radiologist in judging whether the diagnostic is correct or not, and in consequence to accelerate the decision.

Traditional image processing operators often provide some control parameters to tweak the final results. Recently, different convolutional neural networks have been used to approximate or improve these operators. However, in those methods, one single model can only handle one operator of a specific parameter value and does not support parameter tuning. In this paper, we propose a new plugin module, “Adaptive Filterbank Pyramid” , which can be inserted into a backbone network to support multiple operators and continuous parameter tuning. Our module explicitly represents one operator with one filterbank pyramid. To generate the results of a specific operator, the corresponding filterbank pyramid is convolved with the intermediate feature pyramid produced by the backbone network. The weights of the filterbank pyramid are directly regressed by another sub-network, which is jointly trained with the backbone network and adapted to the input parameter, thus enabling continuous parameter tuning. We applied the proposed module for a large variety of image processing tasks, including image smoothing, image denoising, image deblocking, image enhancement and neural style transfer. Experiments show that our method is generalized to different types of image processing tasks and different backbone network structures. Compared to the single-operator-single-parameter baseline, our method can produce comparable results but is significantly more efficient in both training and testing.

Minimally invasive surgical and diagnostic systems are commonly used in clinical practices. However, the accuracy and robustness of these systems depend heavily on computer based processes such as tracking, detecting or segmenting clinically meaningful regions of interest, which are significantly affected by the inherent specular reflections that appear on the organs' surfaces. Restoration of the acquired data for clinical purposes still presents challenges because of the high texture and color variations across the image. In this work, we propose a novel fully-automated solution for endoscopic image restoration, which we call ReTouchImg. Our approach is designed as a two-step scheme. The first is a detection step that is based on the synergy of a set of color variations and gradient information conditions. For the second step, we introduce an inpainting process which is based on graph data structures for recovering the missing information. We exhaustively evaluate our approach on real endoscopic datasets and compare it against some works from the body of literature. We also demonstrate that our solution deals with complex cases such as strong illumination variation and large affected areas through a careful quantitative evaluation of a range of numerical results.

Motion degradation is a central problem in Magnetic Resonance Imaging (MRI). This work addresses the problem of how to obtain higher quality, super-resolved motion-free reconstructions from highly undersampled MRI data. In this work, we present for the first time a variational multi-task framework that allows joining three relevant tasks in MRI: reconstruction, registration and super-resolution. Our framework takes a set of multiple undersampled MR acquisitions corrupted by motion into a novel multi-task optimisation model, which is composed of an L2 fidelity term that allows sharing representation between tasks, super-resolution foundations and hyperelastic deformations to model biological tissue behaviors. We demonstrate that this combination yields significant improvements over sequential models and other bi-task methods. Our results exhibit fine details and compensate for motion producing sharp and highly textured images compared to state of the art methods while keeping low CPU time. Our improvements are appraised on both clinical assessment and statistical analysis.

We consider the problem of segmenting an image into superpixels in the context of k -means clustering, in which we wish to decompose an image into local, homogeneous regions corresponding to the underlying objects. Our novel approach builds upon the widely used Simple Linear Iterative Clustering (SLIC), and incorporate a measure of objects’ structure based on the spectral residual of an image. Based on this combination, we propose a modified initialisation scheme and search metric, which keeps fine-details. This combination leads to better adherence to object boundaries, while preventing unnecessary segmentation of large, uniform areas, and remaining computationally tractable in comparison to other methods. We demonstrate through numerical and visual experiments that our approach outperforms the state-of-the-art techniques.

Semi-supervised learning has received a lot of recent attention as it alleviates the need for large amounts of labelled data which can often be expensive, requires expert knowledge and be time consuming to collect. Recent developments in deep semi-supervised classification have reached unprecedented performance and the gap between supervised and semi-supervised learning is ever-decreasing. This improvement in performance has been based on the inclusion of numerous technical tricks, strong augmentation techniques and costly optimisation schemes with multi-term loss functions. We propose a new framework, LaplaceNet, for deep semi-supervised classification that has a greatly reduced model complexity. We utilise a hybrid approach where pseudolabels are produced by minimising the Laplacian energy on a graph. These pseudo-labels are then used to iteratively train a neural-network backbone. Our model outperforms state-of-the art methods for deep semi-supervised classification, over several benchmark datasets. Furthermore, we consider the application of strong-augmentations to neural networks theoretically and justify the use of a multi-sampling approach for semi-supervised learning. We demonstrate, through rigorous experimentation, that a multi-sampling augmentation approach improves generalisation and reduces the sensitivity of the network to augmentation., Comment: https://ieeexplore.ieee.org/document/9900300

International audience; Reconstructing high-quality magnetic resonance images (MRIs) from undersampled raw data is of great interest from both technical and clinical point of views. To this date, however, it is still a mathematically and computationally challenging problem due to its severe ill-posedness, resulting from the highly undersampled data. Whilst a number of techniques have been presented to improve image reconstruction, they only account for spatio-temporal regularisation, which shows its limitations in several relevant scenarios including dynamic data. In this work, we propose a new mathematical model for the reconstruction of high-quality medical MRI from few measurements. Our proposed approach combines—in a multi-task and hybrid model—the traditional compressed sensing formulation for the reconstruction of dynamic MRI with motion compensation by learning an optical flow approximation. More precisely, we propose to encode the dynamics in the form of an optical flow model that is sparsely represented over a learned dictionary. This has the advantage that ground truth data is not required in the training of the optical flow term. Furthermore, we present an efficient optimisation scheme to tackle the non-convex problem based on an alternating splitting method. We demonstrate the potentials of our approach through an extensive set of numerical results using different datasets and acceleration factors. Our combined approach reaches and outperforms several state-of-the-art techniques for multi-tasking reconstruction and other classic variational reconstruction schemes. Finally, we show the ability of our technique to transfer phantom based knowledge to real datasets.

A central problem in hyperspectral image classification is obtaining high classification accuracy when using a limited amount of labelled data. In this paper we present a novel graph-based framework, which aims to tackle this problem in the presence of large scale data input. Our approach utilises a novel superpixel method, specifically designed for hyperspectral data, to define meaningful local regions in an image, which with high probability share the same classification label. We then extract spectral and spatial features from these regions and use these to produce a contracted weighted graph-representation, where each node represents a region rather than a pixel. Our graph is then fed into a graph-based semi-supervised classifier which gives the final classification. We show that using superpixels in a graph representation is an effective tool for speeding up graphical classifiers applied to hyperspectral images. We demonstrate through exhaustive quantitative and qualitative results that our proposed method produces accurate classifications when an incredibly small amount of labelled data is used. We show that our approach mitigates the major drawbacks of existing approaches, resulting in our approach outperforming several comparative state-of-the-art techniques., 11 pages

We consider the problem of classifying a medical image dataset when we have a limited amounts of labels. This is very common yet challenging setting as labelled data is expensive, time consuming to collect and may require expert knowledge. The current classification go-to of deep supervised learning is unable to cope with such a problem setup. However, using semi-supervised learning, one can produce accurate classifications using a significantly reduced amount of labelled data. Therefore, semi-supervised learning is perfectly suited for medical image classification. However, there has almost been no uptake of semi-supervised methods in the medical domain. In this work, we propose an all-in-one framework for deep semi-supervised classification focusing on graph based approaches, which up to our knowledge it is the first time that an approach with minimal labels has been shown to such an unprecedented scale with medical data. We introduce the concept of hybrid models by defining a classifier as a combination between an energy-based model and a deep net. Our energy functional is built on the Dirichlet energy based on the graph p-Laplacian. Our framework includes energies based on the \(\ell _1\) and \(\ell _2\) norms. We then connected this energy model to a deep net to generate a much richer feature space to construct a stronger graph. Our framework can be set to be adapted to any complex dataset. We demonstrate, through extensive numerical comparisons, that our approach readily compete with fully-supervised state-of-the-art techniques for the applications of Malaria Cells, Mammograms and Chest X-ray classification whilst using only 20% of labels .

Optical flow in heavy rainy scenes is challenging due to the presence of both rain steaks and rain veiling effect, which break the existing optical flow constraints. Concerning this, we propose a deep-learning based optical flow method designed to handle heavy rain. We introduce a feature multiplier in our network that transforms the features of an image affected by the rain veiling effect into features that are less affected by it, which we call veiling-invariant features. We establish a new mapping operation in the feature space to produce streak-invariant features. The operation is based on a feature pyramid structure of the input images, and the basic idea is to preserve the chromatic features of the background scenes while canceling the rain-streak patterns. Both the veiling-invariant and streak-invariant features are computed and optimized automatically based on the the accuracy of our optical flow estimation. Our network is end-to-end, and handles both rain streaks and the veiling effect in an integrated framework. Extensive experiments show the effectiveness of our method, which outperforms the state of the art method and other baseline methods. We also show that our network can robustly maintain good performance on clean (no rain) images even though it is trained under rain image data.

In this paper, we present a graph-based semi-supervised framework for hyperspectral image classification. We first introduce a novel superpixel algorithm based on the spectral covariance matrix representation of pixels to provide a better representation of our data. We then construct a superpixel graph, based on carefully considered feature vectors, before performing classification. We demonstrate, through a set of experimental results using two benchmarking datasets, that our approach outperforms three state-of-the-art classification frameworks, especially when an extremely small amount of labelled data is used., Comment: Four pages with two figures

This work addresses a central topic in Magnetic Resonance Imaging (MRI) which is the motion-correction problem in a joint reconstruction and registration framework. From a set of multiple MR acquisitions corrupted by motion, we aim at - jointly - reconstructing a single motion-free corrected image and retrieving the physiological dynamics through the deformation maps. To this purpose, we propose a novel variational model. First, we introduce an $L^2$ fidelity term, which intertwines reconstruction and registration along with the weighted total variation. Second, we introduce an additional regulariser which is based on the hyperelasticity principles to allow large and smooth deformations. We demonstrate through numerical results that this combination creates synergies in our complex variational approach resulting in higher quality reconstructions and a good estimate of the breathing dynamics. We also show that our joint model outperforms in terms of contrast, detail and blurring artefacts, a sequential approach., 12 pages, 3 figure, accepted for publication in Scale Space and Variational Methods in Computer Vision conference proceedings

The task of classifying X-ray data is a problem of both theoretical and clinical interest. Whilst supervised deep learning methods rely upon huge amounts of labelled data, the critical problem of achieving a good classification accuracy when an extremely small amount of labelled data is available has yet to be tackled. In this work, we introduce a novel semi-supervised framework for X-ray classification which is based on a graph-based optimisation model. To the best of our knowledge, this is the first method that exploits graph-based semi-supervised learning for X-ray data classification. Furthermore, we introduce a new multi-class classification functional with carefully selected class priors which allows for a smooth solution that strengthens the synergy between the limited number of labels and the huge amount of unlabelled data. We demonstrate, through a set of numerical and visual experiments, that our method produces highly competitive results on the ChestX-ray14 data set whilst drastically reducing the need for annotated data.

Indirect image registration is a promising technique to improve image reconstruction quality by providing a shape prior for the reconstruction task. In this paper, we propose a novel hybrid method that seeks to reconstruct high quality images from few measurements whilst requiring low computational cost. With this purpose, our framework intertwines indirect registration and reconstruction tasks is a single functional. It is based on two major novelties. Firstly, we introduce a model based on deep nets to solve the indirect registration problem, in which the inversion and registration mappings are recurrently connected through a fixed-point interaction based sparse optimisation. Secondly, we introduce specific inversion blocks, that use the explicit physical forward operator, to map the acquired measurements to the image reconstruction. We also introduce registration blocks based deep nets to predict the registration parameters and warp transformation accurately and efficiently. We demonstrate, through extensive numerical and visual experiments, that our framework outperforms significantly classic reconstruction schemes and other bi-task method; this in terms of both image quality and computational time. Finally, we show generalisation capabilities of our approach by demonstrating their performance on fast Magnetic Resonance Imaging (MRI), sparse view computed tomography (CT) and low dose CT with measurements much below the Nyquist limit.

International audience; We address the problem of reconstructing high quality images from undersampled MRI data. This is a challenging task due to the highly ill-posed nature of the problem. In particular, in dynamic MRI scans, the interaction between the target structure and the physical motion affects the acquired measurements leading to blurring artefacts and loss of fine details. In this work, we propose a framework for dynamic MRI reconstruction framed under a new multi-task optimisation model called Compressed Sensing Plus Motion (CS+M). Firstly, we propose a single optimisation problem that simultaneously computes the MRI reconstruction and the physical motion. Secondly, we show our model can be efficiently solved by breaking it up into two more computationally tractable problems. The potentials and generalisation capabilities of our approach are demonstrated in different clinical applications including cardiac cine, cardiac perfusion and brain perfusion imaging. We show, through numerical and graphical experiments, that the proposed scheme reduces blurring artefacts and preserves the target shape and fine details. We also report the highest quality reconstruction under highly undersampling rates in comparison to several state of the art techniques.

This paper addresses the search for a fast and meaningful image segmentation in the context of k-means clustering. The proposed method builds on a widely-used local version of Lloyd's algorithm, called Simple Linear Iterative Clustering (SLIC). We propose an algorithm which extends SLIC to dynamically adjust the local search, adopting superpixel resolution dynamically to structure existent in the image, and thus provides for more meaningful superpixels in the same linear runtime as standard SLIC. The proposed method is evaluated against state-of-the-art techniques and improved boundary adherence and undersegmentation error are observed, whilst still remaining among the fastest algorithms which are tested.

Robotic-assisted surgeries are commonly used today as a more efficient alternative to traditional surgical options. Both surgeons and patients benefit from those systems, as they offer many advantages, including less trauma and blood loss, fewer complications, and better ergonomics. However, a remaining limitation of currently available surgical systems is the lack of force feedback due to the teleoperation setting, which prevents direct interaction with the patient. Once the force information is obtained by either a sensing device or indirectly through vision-based force estimation, a concern arises on how to transmit this information to the surgeon. An attractive alternative is sensory substitution, which allows transcoding information from one sensory modality to present it in a different sensory modality. In the current work, we used visual feedback to convey interaction forces to the surgeon. Our overarching goal was to address the following question: How should interaction forces be displayed to support efficient comprehension by the surgeon without interfering with the surgeon’s perception and workflow during surgery? Until now, the use the visual modality for force feedback has not been carefully evaluated. For this reason, we conducted an experimental study with two aims: (1) to demonstrate the potential benefits of using this modality and (2) to understand the surgeons’ perceptual preferences. The results derived from our study of 28 surgeons revealed a strong positive acceptance of the users (96%) using this modality. Moreover, we found that for surgeons to easily interpret the information, their mental model must be considered, meaning that the design of the visualizations should fit the perceptual and cognitive abilities of the end user. To our knowledge, this is the first time that these principles have been analyzed for exploring sensory substitution in medical robotics. Finally, we provide user-centered recommendations for the design of visual displays for robotic surgical systems.

Removing reflection artefacts from a single image is a problem of both theoretical and practical interest, which still presents challenges because of the massively ill-posed nature of the problem. In this paper, we propose a technique based on a novel optimization problem. First, we introduce a simple user interaction scheme, which helps minimize information loss in the reflection-free regions. Second, we introduce an H 2 fidelity term, which preserves fine detail while enforcing the global color similarity. We show that this combination allows us to mitigate the shortcomings in structure and color preservation, which presents some of the most prominent drawbacks in the existing methods for reflection removal. We demonstrate, through numerical and visual experiments, that our method is able to outperform the state-of-the-art model-based methods and compete with recent deep-learning approaches.

Technical advancements have been part of modern medical solutions as they promote better surgical alternatives that serve to the benefit of patients. Particularly with cardiovascular surgeries, robotic surgical systems enable surgeons to perform delicate procedures on a beating heart, avoiding the complications of cardiac arrest. This advantage comes with the price of having to deal with a dynamic target which presents technical challenges for the surgical system. In this work, we propose a solution for cardiac motion estimation. Our estimation approach uses a variational framework that guarantees preservation of the complex anatomy of the heart. An advantage of our approach is that it takes into account different disturbances, such as specular reflections and occlusion events. This is achieved by performing a preprocessing step that eliminates the specular highlights and a predicting step, based on a conditional restricted Boltzmann machine, that recovers missing information caused by partial occlusions. We carried out exhaustive experimentations on two datasets, one from a phantom and the other from an in vivo procedure. The results show that our visual approach reaches an average minima in the order of magnitude of $$10^{-7}$$ while preserving the heart’s anatomical structure and providing stable values for the Jacobian determinant ranging from 0.917 to 1.015. We also show that our specular elimination approach reaches an accuracy of 99% compared to a ground truth. In terms of prediction, our approach compared favorably against two well-known predictors, NARX and EKF, giving the lowest average RMSE of 0.071. Our approach avoids the risks of using mechanical stabilizers and can also be effective for acquiring the motion of organs other than the heart, such as the lung or other deformable objects.