Federated learning can support multiple clients to train a global model with the assistance of the central server without sharing raw clients' data. However, the gradients uploaded by clients still compromise privacy and federated learning is vulnerable to poisoning attacks by malicious clients. Previous studies have focused on privacy preservation or defense against poisoning attacks, respectively. In practice, privacy preservation further increases the difficulty of defending against poisoning attacks, leading to the problem of severe degradation of model accuracy under strong privacy preservation. In this paper, we propose a lightweight, robust, and privacy-preserving federated learning scheme that can effectively resist poisoning attacks under strong privacy protection and prevent model accuracy degradation. We employ a differential privacy technique that adds noise to the upload gradient to protect data privacy. To address the problem that it is difficult to detect malicious clients under strong privacy preservation, we design a probabilistic grouping mechanism based on trust scores to support the detection of malicious clients. We conduct extensive experiments to evaluate our scheme, and the results show that our scheme can resist different types of poisoning attacks under strong privacy preservation, thus improving the accuracy of the model.

Endoscopes, both white light and fluorescence, face challenges like insufficient illumination during surgical procedures. Deep learning, especially diffusion models, demonstrates considerable potential for low-light enhancement in the medical field. However, challenges in enhancing endoscopic images persist, including high computational resource consumption, lengthy processing times, and potential result distortion. To address these issues, we propose a wavelet-based diffusion method for fast and efficient low-light surgical endoscopic image enhancement, dubbed WDiff. It utilizes wavelet transform to reduce computational resource and enhance inference speed while preserving key features. To avoid degradation during reverse process, we introduce a Degradation Corrector Unit (DCU) to ensure accurate sampling results. Moreover, we design a Detail Coefficients Restoration Block (DCRB) to reconstruct the local sparse information in detail coefficients across horizontal, vertical, and diagonal orientations. Extensive experiments on benchmark datasets demonstrate that our method outperforms other SOTA methods both visually and quantitatively, achieving an optimal balance between complexity and efficiency.

High dynamic range (HDR) imaging involves capturing a series of frames of the same scene, each with different exposure settings, to broaden the dynamic range of light. This can be achieved through burst capturing or using staggered HDR sensors that capture long and short exposures simultaneously in the camera image signal processor (ISP). Within camera ISP pipeline, illuminant estimation is a crucial step aiming to estimate the color of the global illuminant in the scene. This estimation is used in camera ISP white-balance module to remove undesirable color cast in the final image. Despite the multiple frames captured in the HDR pipeline, conventional illuminant estimation methods often rely only on a single frame of the scene. In this paper, we explore leveraging information from frames captured with different exposure times. Specifically, we introduce a simple feature extracted from dual-exposure images to guide illuminant estimators, referred to as the dual-exposure feature (DEF). To validate the efficiency of DEF, we employed two illuminant estimators using the proposed DEF: 1) a multilayer perceptron network (MLP), referred to as exposure-based MLP (EMLP), and 2) a modified version of the convolutional color constancy (CCC) to integrate our DEF, that we call ECCC. Both EMLP and ECCC achieve promising results, in some cases surpassing prior methods that require hundreds of thousands or millions of parameters, with only a few hundred parameters for EMLP and a few thousand parameters for ECCC.

Multiscale features are of great importance in modern convolutional neural networks, showing consistent performance gains on numerous vision tasks. Therefore, many plug-and-play blocks are introduced to upgrade existing convolutional neural networks for stronger multiscale representation ability. However, the design of plug-and-play blocks is getting more and more complex, and these manually designed blocks are not optimal. In this work, we propose PP-NAS to develop plug-and-play blocks based on neural architecture search (NAS). Specifically, we design a new search space PPConv and develop a search algorithm consisting of one-level optimization, zero-one loss, and connection existence loss. PP-NAS minimizes the optimization gap between super-net and subarchitectures and can achieve good performance even without retraining. Extensive experiments on image classification, object detection, and semantic segmentation verify the superiority of PP-NAS over state-of-the-art CNNs (e.g., ResNet, ResNeXt, and Res2Net). Our code is available at https://github.com/ainieli/PP-NAS.

We consider the problem of stragglers in distributed computing systems. Stragglers, which are compute nodes that unpredictably slow down, often increase the completion times of tasks. One common approach to mitigating stragglers is work replication, where only the first completion among replicated tasks is accepted, discarding the others. However, discarding work leads to resource wastage. In this article, we propose a method for exploiting the work completed by stragglers rather than discarding it. The idea is to increase the granularity of the assigned work, and to increase the frequency of worker updates. We show that the proposed method reduces the completion time of tasks via experiments performed on a simulated cluster as well as on Amazon EC2 with Apache Hadoop.

Traffic processing on cloud-scale bandwidths has given rise to a new type of network structure, comprising a large number of highly-structured virtual entities working in close harmony. This structure, which we call a virtual sub-network, might be in need of migration, for reasons of load-balancing, maintenance, and disaster prevention. In this paper, we argue that the common migration schemes are not adequate for the complexity of this task. Therefore, we present Qanat, a migration system specifically optimized for the live migration of a virtual sub-network in its entirety to a different physical location. We show how Qanat employs widely-used techniques, such as traffic prioritization, buffering, and network tunnels, to overcome the main issues of live migration. In the paper, we categorize the main challenges of the migration task, provide an analytical study of Qanat’s algorithms, and measure its performance metrics through large-scale simulations. We conclude that Qanat can efficiently and transparently migrate virtual sub-networks and can provide a useful tool for system administrators.

Automatic data augmentation is a technique to automatically search for strategies for image transformations, which can improve the performance of different vision tasks. RandAugment (RA), one of the most widely used automatic data augmentations, achieves great success in different scales of models and datasets. However, RA randomly selects transformations with equivalent probabilities and applies a single magnitude for all transformations, which is suboptimal for different models and datasets. In this paper, we develop Differentiable RandAugment (DRA) to learn selecting weights and magnitudes of transformations for RA. The magnitude of each transformation is modeled following a normal distribution with both learnable mean and standard deviation. We also introduce the gradient of transformations to reduce the bias in gradient estimation and KL divergence as part of the loss to reduce the optimization gap. Experiments on CIFAR-10/100 and ImageNet demonstrate the efficiency and effectiveness of DRA. Searching for only 0.95 GPU hours on ImageNet, DRA can reach a Top-1 accuracy of 78.19% with ResNet-50, which outperforms RA by 0.28% under the same settings. Transfer learning on object detection also demonstrates the power of DRA. The proposed DRA is one of the few that surpasses RA on ImageNet and has great potential to be integrated into modern training pipelines to achieve state-of-the-art performance. Our code will be made publicly available for out-of-the-box use.

Glioma grading during surgery can help clinical treatment planning and prognosis, but intraoperative pathological examination of frozen sections is limited by the long processing time and complex procedures. Near-infrared fluorescence imaging provides chances for fast and accurate real-time diagnosis. Recently, deep learning techniques have been actively explored for medical image analysis and disease diagnosis. However, issues of near-infrared fluorescence images, including small-scale, noise, and low-resolution, increase the difficulty of training a satisfying network. Multi-modal imaging can provide complementary information to boost model performance, but simultaneously designing a proper network and utilizing the information of multi-modal data is challenging. In this work, we propose a novel neural architecture search method DLS-DARTS to automatically search for network architectures to handle these issues. DLS-DARTS has two learnable stems for multi-modal low-level feature fusion and uses a modified perturbation-based derivation strategy to improve the performance on the area under the curve and accuracy. White light imaging and fluorescence imaging in the first near-infrared window (650-900 nm) and the second near-infrared window (1,000-1,700 nm) are applied to provide multi-modal information on glioma tissues. In the experiments on 1,115 surgical glioma specimens, DLS-DARTS achieved an area under the curve of 0.843 and an accuracy of 0.634, which outperformed manually designed convolutional neural networks including ResNet, PyramidNet, and EfficientNet, and a state-of-the-art neural architecture search method for multi-modal medical image classification. Our study demonstrates that DLS-DARTS has the potential to help neurosurgeons during surgery, showing high prospects in medical image analysis.

Scalability is an important property for a resource manager in the cloud. It is challenging to manage large-scale cluster resources while serving a large number of requests. Moreover, cloud-based applications and new technologies result in changeable requirements, so a cloud resource manager needs to continuously evolve. This paper presents a novel architecture design for a scalable and extensible resource manager. Our resource manager named Fundy employs a microservices architecture based on an in-memory data grid. The in-memory data grid accelerates data processing and enables Fundy to easily scale out. The data grid facilitates Fundy's microservices-based architecture. Because of the architecture, it is possible to rapidly deploy new features in Fundy. Fundy also enables multiple schedulers to schedule different workloads on shared infrastructure. In addition, we introduce a new packing algorithm to improve the allocation quality of our scheduler and a tensor scheduling algorithm to speed up parallel processing of requests by orders of magnitude. Fundy has been deployed in our production public cloud. This paper includes evaluation based on real-world traces and the results highlight the advantages of Fundy.

Load migration protocols have been used for load balancing in network controllers. In this poster, we argue that other network applications (e.g., power saving, network security, failure recovery, etc.) have properties that might require different load migration protocols. We introduce four new load migration protocols and show how they might match different application requirements better. We present preliminary experimental results for one of these protocols that show more than 20%-30% speedup in the total load migration time.

As network traffic and conditions change, the load on different instances of control plane changes. To ensure various control applications can operate continuously and efficiently, we need to migrate the load among controller instances. For this, we need a migration schedule that minimizes the overall migration time while ensuring the quality of service and controller resource constraints. In this poster, we show this problem is NP-hard, and show how a heuristic algorithm performs close to the best existing solution with orders of magnitude reduction in scheduling time.

OBJECTIVE: The coronavirus disease 2019 (COVID-19) is rapidly spreading inside China and internationally. We aimed to construct a model integrating information from radiomics and deep learning (DL) features to discriminate critical cases from severe cases of COVID-19 using computed tomography (CT) images. METHODS: We retrospectively enrolled 217 patients from three centers in China, including 82 patients with severe disease and 135 with critical disease. Patients were randomly divided into a training cohort (n = 174) and a test cohort (n = 43). We extracted 102 3-dimensional radiomic features from automatically segmented lung volume and selected the significant features. We also developed a 3-dimensional DL network based on center-cropped slices. Using multivariable logistic regression, we then created a merged model based on significant radiomic features and DL scores. We employed the area under the receiver operating characteristic curve (AUC) to evaluate the model's performance. We then conducted cross validation, stratified analysis, survival analysis, and decision curve analysis to evaluate the robustness of our method. RESULTS: The merged model can distinguish critical patients with AUCs of 0.909 (95% confidence interval [CI]: 0.859-0.952) and 0.861 (95% CI: 0.753-0.968) in the training and test cohorts, respectively. Stratified analysis indicated that our model was not affected by sex, age, or chronic disease. Moreover, the results of the merged model showed a strong correlation with patient outcomes. SIGNIFICANCE: A model combining radiomic and DL features of the lung could help distinguish critical cases from severe cases of COVID-19.

Fluorescence molecular tomography (FMT), which can visualize the distribution of fluorescence biomarkers, has become a novel three-dimensional noninvasive imaging technique for in vivo studies such as tumor detection and lymph node location. However, it remains a challenging problem to achieve satisfactory reconstruction performance of conventional FMT in the first near-infrared window (NIR-I, 700-900nm) because of the severe scattering of NIR-I light. In this study, a promising FMT method for heterogeneous mice was proposed to improve the reconstruction accuracy using the second near-infrared window (NIR-II, 1000-1700nm), where the light scattering significantly reduced compared with NIR-I. The optical properties of NIR-II were analyzed to construct the forward model for NIR-II FMT. Furthermore, to raise the accuracy of solution of the inverse problem, we proposed a novel Gaussian weighted neighborhood fused Lasso (GWNFL) method. Numerical simulation was performed to demonstrate the outperformance of GWNFL compared with other algorithms. Besides, a novel NIR-II/NIR-I dual-modality FMT system was developed to contrast the in vivo reconstruction performance between NIR-II FMT and NIR-I FMT. To compare the reconstruction performance of NIR-II FMT with traditional NIR-I FMT, numerical simulations and in vivo experiments were conducted. Both the simulation and in vivo results showed that NIR-II FMT outperformed NIR-I FMT in terms of location accuracy and spatial overlap index. It is believed that this study could promote the development and biomedical application of NIR-II FMT in the future.

Cerenkov luminescence tomography (CLT) is a promising imaging tool for obtaining three-dimensional (3D) non-invasive visualization of the in vivo distribution of radiopharmaceuticals. However, the reconstruction performance remains unsatisfactory for biomedical applications because the inverse problem of CLT is severely ill-conditioned and intractable. In this study, therefore, a novel non-negative iterative convex refinement (NNICR) approach was utilized to improve the CLT reconstruction accuracy, robustness as well as the shape recovery capability. The spike and slab prior information was employed to capture the sparsity of Cerenkov source, which could be formalized as a non-convex optimization problem. The NNICR approach solved this non-convex problem by refining the solutions of the convex sub-problems. To evaluate the performance of the NNICR approach, numerical simulations and in vivo tumor-bearing mice models experiments were conducted. Conjugated gradient based Tikhonov regularization approach (CG-Tikhonov), fast iterative shrinkage-thresholding algorithm based Lasso approach (Fista-Lasso) and Elastic-Net regularization approach were used for the comparison of the reconstruction performance. The results of these experiments demonstrated that the NNICR approach obtained superior reconstruction performance in terms of location accuracy, shape recovery capability, robustness and in vivo practicability. It was believed that this study would facilitate the preclinical and clinical applications of CLT in the future.

Cerenkov luminescence tomography (CLT) provides a novel technique for 3-D noninvasive detection of radiopharmaceuticals in living subjects. However, because of the severe scattering of Cerenkov light, the reconstruction accuracy and stability of CLT is still unsatisfied. In this paper, a modified weight multispectral CLT (wmCLT) reconstruction strategy was developed which split the Cerenkov radiation spectrum into several sub-spectral bands and weighted the sub-spectral results to obtain the final result. To better evaluate the property of the wmCLT reconstruction strategy in terms of accuracy, stability and practicability, several numerical simulation experiments and in vivo experiments were conducted and the results obtained were compared with the traditional multispectral CLT (mCLT) and hybrid-spectral CLT (hCLT) reconstruction strategies. The numerical simulation results indicated that wmCLT strategy significantly improved the accuracy of Cerenkov source localization and intensity quantitation and exhibited good stability in suppressing noise in numerical simulation experiments. And the comparison of the results achieved from different in vivo experiments further indicated significant improvement of the wmCLT strategy in terms of the shape recovery of the bladder and the spatial resolution of imaging xenograft tumors. Overall the strategy reported here will facilitate the development of nuclear and optical molecular tomography in theoretical study.

Heart ventricular mechanics has been investigated intensively in the last four decades. The passive material properties, the ventricular geometry and muscular architecture, and the myocardial activation are among the most important determinants of cardiac mechanics. The heart muscle is anisotropic, inhomogeneous, and highly nonlinear. The heart ventricular geometry is irregular and object dependent. The muscular architecture includes the organization of the fiber and the connective tissues. Studies of the myocardial activation have been carried out at both cell and tissue levels.Previous work from our research group has successfully estimated the in-vivo motion and deformation of both the left and the right ventricles. In this paper, we present an iterative model to estimate the in-vivo myocardium material properties, the active forces generated along fiber orientation, and strain and stress distribution in both ventricles. Compared to the strain energy function approach, our model is more intuitively understandable. Using the model, we have simulated the mechanical events of a few different heart diseases. Noticeable strain and stress differences are found between normal and diseased hearts.