Search

We propose a framework that combines traditional, hand-crafted algorithms and recent advances in deep learning to obtain high-quality, high-resolution disparity maps from stereo images. By casting the refinement process as a continuous feature sampling strategy, our neural disparity refinement network can estimate an enhanced disparity map at any output resolution. Our solution can process any disparity map produced by classical stereo algorithms, as well as those predicted by modern stereo networks or even different depth-from-images approaches, such as the COLMAP structure-from-motion pipeline. Nonetheless, when deployed in the former configuration, our framework performs at its best in terms of zero-shot generalization from synthetic to real images. Moreover, its continuous formulation allows for easily handling the unbalanced stereo setup very diffused in mobile phones.

Immunoprecipitation (IP) coupled with mass spectrometry effectively maps protein-protein interactions when genome-wide, affinity-tagged cell collections are used. Such studies have recorded significant portions of the compositions of physiological protein complexes, providing draft 'interactomes'; yet many constituents of protein complexes still remain uncharted. This gap exists partly because high-throughput approaches cannot optimize each IP. A key challenge for IP optimization is stabilizing in vivo interactions during the transfer from cells to test tubes; failure to do so leads to the loss of genuine interactions during the IP and subsequent failure to detect. Our high-content screening method explores the relationship between in vitro chemical conditions and IP outcomes, enabling rapid empirical optimization of conditions for capturing target macromolecular assemblies.

Objectives: To analyze the incidence and predictive factors of postoperative acute kidney injury (AKI) after elective standard endovascular aortic repair (EVAR) in a large recent, multicenter cohort., Materials and Methods: This is a multicenter, retrospective, financially unsupported physician-initiated observational cohort study. Between January 2018 and March 2021, only patients treated with elective standard EVAR for infrarenal non-infected abdominal aortic aneurysm were analyzed. Patients already on hemodialysis (HD) were excluded. AKI was defined as an increase in serum creatinine (sCr) ≥0.3 mg/dL within 48 hours or an increase in sCr to ≥1.5 times baseline known or presumed to have occurred within 7 days, or a urine volume of <0.5 mL/kg/h for 6 hours. Primary outcomes of interest were AKI incidence at 30 days and freedom from HD at 1-year follow-up. Secondary outcomes were freedom from severe postoperative complication, and freedom from aorta-related mortality (ARM) at 1 year., Results: The final cohort analyzed 526 (29.8%). There were 489 (93%) males and 37 (7%) females: the median age was 76 years (interquartile range [IQR], 71-81). Chronic kidney disease (CKD) was present in 86 (16.3%) patients. Early mortality was observed in 8 (1.5%) patients, none was aorta-related. Complication rate was 17.1% (n=89). AKI was observed in 17 (3.2%). Renal replacement therapy was needed in 4 (0.8%). HD was transitory in 2 cases and definitive in 1. Binary logistic regression analysis identified CKD (odds ratio [OR]: 4.68, 95% confidence interval [CI]: 2.10-10.45, p<0.001), and the presence of renal artery stenosis (OR: 3.80, 95% CI: 1.35-10.66, p=0.011) to be associated with an increased risk of postoperative AKI. Estimated freedom from ARM was 94.9% at 1 year. Estimated freedom from HD rate at 1 year was 94%: This was significantly different between patients with preoperative CKD and those who did not have preoperative CKD (log-rank, p=0.042)., Conclusion: AKI after elective standard EVAR still occurs but with an acceptably low incidence rate. Preoperative CKD is the most important predictor for postoperative AKI, which was not associated with the need for HD at 1-year follow-up but with a higher propensity of mortality., Clinical Impact: This "real world" experience confirm that EVAR performed with standard contrats agent protocol remains safe for acute kidney injury development. Therefore, only patients presenting with preoperative borderline or ascertained chronic kidney disease will take the most advantage from the use of carbon dioxide contrast., Competing Interests: Declaration of Conflicting InterestsThe author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Estimating depth from images nowadays yields outstanding results, both in terms of in-domain accuracy and generalization. However, we identify two main challenges that remain open in this field: dealing with non-Lambertian materials and effectively processing high-resolution images. Purposely, we propose a novel dataset that includes accurate and dense ground-truth labels at high resolution, featuring scenes containing several specular and transparent surfaces. Our acquisition pipeline leverages a novel deep space-time stereo framework, enabling easy and accurate labeling with sub-pixel precision. The dataset is composed of 606 samples collected in 85 different scenes, each sample includes both a high-resolution pair (12 Mpx) as well as an unbalanced stereo pair (Left: 12 Mpx, Right: 1.1 Mpx), typical of modern mobile devices that mount sensors with different resolutions. Additionally, we provide manually annotated material segmentation masks and 15 K unlabeled samples. The dataset is composed of a train set and two test sets, the latter devoted to the evaluation of stereo and monocular depth estimation networks. Our experiments highlight the open challenges and future research directions in this field.

Estimating treatment effects conditional on observed covariates can improve the ability to tailor treatments to particular individuals. Doing so effectively requires dealing with potential confounding, and also enough data to adequately estimate effect moderation. A recent influx of work has looked into estimating treatment effect heterogeneity using data from multiple randomized controlled trials and/or observational datasets. With many new methods available for assessing treatment effect heterogeneity using multiple studies, it is important to understand which methods are best used in which setting, how the methods compare to one another, and what needs to be done to continue progress in this field. This paper reviews these methods broken down by data setting: aggregate-level data, federated learning, and individual participant-level data. We define the conditional average treatment effect and discuss differences between parametric and nonparametric estimators, and we list key assumptions, both those that are required within a single study and those that are necessary for data combination. After describing existing approaches, we compare and contrast them and reveal open areas for future research. This review demonstrates that there are many possible approaches for estimating treatment effect heterogeneity through the combination of datasets, but that there is substantial work to be done to compare these methods through case studies and simulations, extend them to different settings, and refine them to account for various challenges present in real data.

Availability of labelled data is the major obstacle to the deployment of deep learning algorithms for computer vision tasks in new domains. The fact that many frameworks adopted to solve different tasks share the same architecture suggests that there should be a way of reusing the knowledge learned in a specific setting to solve novel tasks with limited or no additional supervision. In this work, we first show that such knowledge can be shared across tasks by learning a mapping between task-specific deep features in a given domain. Then, we show that this mapping function, implemented by a neural network, is able to generalize to novel unseen domains. Besides, we propose a set of strategies to constrain the learned feature spaces, to ease learning and increase the generalization capability of the mapping network, thereby considerably improving the final performance of our framework. Our proposal obtains compelling results in challenging synthetic-to-real adaptation scenarios by transferring knowledge between monocular depth estimation and semantic segmentation tasks.

Correspondences between 3D keypoints generated by matching local descriptors are a key step in 3D computer vision and graphic applications. Learned descriptors are rapidly evolving and outperforming the classical handcrafted approaches in the field. Yet, to learn effective representations they require supervision through labeled data, which are cumbersome and time-consuming to obtain. Unsupervised alternatives exist, but they lag in performance. Moreover, invariance to viewpoint changes is attained either by relying on data augmentation, which is prone to degrading upon generalization on unseen datasets, or by learning from handcrafted representations of the input which are already rotation invariant but whose effectiveness at training time may significantly affect the learned descriptor. We show how learning an equivariant 3D local descriptor instead of an invariant one can overcome both issues. LEAD (Local EquivAriant Descriptor) combines Spherical CNNs to learn an equivariant representation together with plane-folding decoders to learn without supervision. Through extensive experiments on standard surface registration datasets, we show how our proposal outperforms existing unsupervised methods by a large margin and achieves competitive results against the supervised approaches, especially in the practically very relevant scenario of transfer learning.

State-of-the-art approaches to infer dense depth measurements from images rely on CNNs trained end-to-end on a vast amount of data. However, these approaches suffer a drastic drop in accuracy when dealing with environments much different in appearance and/or context from those observed at training time. This domain shift issue is usually addressed by fine-tuning on smaller sets of images from the target domain annotated with depth labels. Unfortunately, relying on such supervised labeling is seldom feasible in most practical settings. Therefore, we propose an unsupervised domain adaptation technique which does not require groundtruth labels. Our method relies only on image pairs and leverages on classical stereo algorithms to produce disparity measurements alongside with confidence estimators to assess upon their reliability. We propose to fine-tune both depth-from-stereo as well as depth-from-mono architectures by a novel confidence-guided loss function that handles the measured disparities as noisy labels weighted according to the estimated confidence. Extensive experimental results based on standard datasets and evaluation protocols prove that our technique can address effectively the domain shift issue with both stereo and monocular depth prediction architectures and outperforms other state-of-the-art unsupervised loss functions that may be alternatively deployed to pursue domain adaptation.

Camera pose estimation is an important problem in computer vision, with applications as diverse as simultaneous localisation and mapping, virtual/augmented reality and navigation. Common techniques match the current image against keyframes with known poses coming from a tracker, directly regress the pose, or establish correspondences between keypoints in the current image and points in the scene in order to estimate the pose. In recent years, regression forests have become a popular alternative to establish such correspondences. They achieve accurate results, but have traditionally needed to be trained offline on the target scene, preventing relocalisation in new environments. Recently, we showed how to circumvent this limitation by adapting a pre-trained forest to a new scene on the fly. The adapted forests achieved relocalisation performance that was on par with that of offline forests, and our approach was able to estimate the camera pose in close to real time, which made it desirable for systems that require online relocalisation. In this paper, we present an extension of this work that achieves significantly better relocalisation performance whilst running fully in real time. To achieve this, we make several changes to the original approach: (i) instead of simply accepting the camera pose hypothesis produced by RANSAC without question, we make it possible to score the final few hypotheses it considers using a geometric approach and select the most promising one; (ii) we chain several instantiations of our relocaliser (with different parameter settings) together in a cascade, allowing us to try faster but less accurate relocalisation first, only falling back to slower, more accurate relocalisation as necessary; and (iii) we tune the parameters of our cascade, and the individual relocalisers it contains, to achieve effective overall performance. Taken together, these changes allow us to significantly improve upon the performance our original state-of-the-art method was able to achieve on the well-known 7-Scenes and Stanford 4 Scenes benchmarks. As additional contributions, we present a novel way of visualising the internal behaviour of our forests, and use the insights gleaned from this to show how to entirely circumvent the need to pre-train a forest on a generic scene.

Pipelines to recognize 3D objects despite clutter and occlusions usually end up with a final verification stage whereby recognition hypotheses are validated or dismissed based on how well they explain sensor measurements. Unlike previous work, we propose a Global Hypothesis Verification (GHV) approach which regards all hypotheses jointly so as to account for mutual interactions. GHV provides a principled framework to tackle the complexity of our visual world by leveraging on a plurality of recognition paradigms and cues. Accordingly, we present a 3D object recognition pipeline deploying both global and local 3D features as well as shape and color. Thereby, and facilitated by the robustness of the verification process, diverse object hypotheses can be gathered and weak hypotheses need not be suppressed too early to trade sensitivity for specificity. Experiments demonstrate the effectiveness of our proposal, which significantly improves over the state-of-art and attains ideal performance (no false negatives, no false positives) on three out of the six most relevant and challenging benchmark datasets.

Silicon nanoparticles obtained by ball-milling of a 50% porosity silicon layer have been optically trapped when dispersed in a water-surfactant environment. We measured the optical force constants using linearly and radially polarized trapping beams finding a reshaping of the optical potential and an enhanced axial spring constant for the latter. These measurements open perspectives for the control and handling of silicon nanoparticles as labeling agents in biological analysis and fluorescence imaging techniques.

Introduction: Myelodysplastic syndromes (MDS), characterized by ineffective hematopoiesis and dysplasia in one or more lineages, produce life-threatening cytopenias and progress to acute myeloid leukemia (AML). Growing evidence suggests that targeting epigenetic mechanisms improves MDS/AML pathophysiology., Areas Covered: This review provides an understanding of studies investigating novel agents published up to January 2011 aimed at normalizing and monitoring the epigenetic profile of the MDS cancer cell. The authors discuss how non-intensive epigenetic therapy can 're-programme' gene expression patterns of abnormal hematopoiesis in MDS. Recently FDA-approved DNA-methyltransferase inhibitors, 5-azacytidine and 5-aza-2'-deoxycytidine or decitabine, represent frontline nonablative treatments, while combinations with histone deacetylase inhibitors show promising synergism in preclinical and Phase I/II trials in tumor suppressor gene re-expression and overall survival. Additional epigenetic mechanisms including non-encoding transcripts with inhibitory posttranscriptional regulatory functions, such as microRNAs, though not fully understood, present novel molecular and clinical implications in these disorders., Expert Opinion: Alongside current single-agent epigenetic regimens, combination therapies represent potentially effective options for intermediate-2 and high-risk MDS. Methylation profiles and gene mutation predictors provide promising areas of development for monitoring MDS disease progression and outcome, while targeting microRNA dysregulation represents an important therapeutic goal.

To improve developmental competence of in vitro matured oocytes, culture medium can be supplemented with hypoxanthine (Hx) and FSH or epidermal growth factor (EGF) to trigger the activation of essential signalling pathways regulating meiotic resumption and progression. Since the serine/threonine kinase, Akt, contributes to the regulation of the meiotic cell cycle, this study analysed its expression level and localization at the meiotic spindle in oocytes matured in vivo or in vitro in the presence of Hx-FSH or Hx-EGF. Independently of culture conditions adopted, Akt mRNA concentration did not vary from germinal vesicle to metaphase I (MI), while at MII a significant decrease in Akt1 mRNA concentration was recorded in oocytes matured in vivo and in those stimulated by Hx-EGF (P < 0.05). Phoshorylated Akt protein content was similar in the different groups of MI oocytes, but it decreased at MII in oocytes matured either in vivo or in vitro with Hx-EGF. Ser-473-phosphorylated Akt was localized uniformly to the meiotic spindle in more than 90% of oocytes. These results indicate that, in mouse oocytes, Akt expression is differentially regulated during in vivo and in vitro maturation and suggest that EGF could be a positive modulator, even stronger than FSH, of oocyte meiotic maturation., (Copyright (c) 2009 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved.)

The protein-ligand molecular interactions imply strong geometrical and structural rearrangements of the biological complex which are normally detected by high sensitivity optical techniques such as time-resolved fluorescence microscopy. In this work, we have measured, by optical spectroscopic reflectometry in the visible-near-infrared region, the interaction between a sugar binding protein (SBP), covalently bound on the surface of a porous silicon (PSi) microcavity, and glucose, at different concentrations and temperatures. Variable-angle spectroscopic ellipsometric (VASE) characterization of protein-functionalized PSi layers confirms that the protein-ligand system has an overall volume smaller than the SBP alone.