Search for: All records

This paper proposes a data-driven adaptive coordination of damping controllers to enhance power system stability. The coordination uses wide-area frequency measurements to select the switching status (on/off) of damping controllers (DC) enabled in electronically-interfaced resources (EIR). This is done by using the total action (TA), a dynamic performance measure of the oscillation energy related to the synchronous generators; and deep neural networks (DNNs), a powerful learning algorithm capable of providing accurate model regression between the grid measurements and the TA. The concept is tested in the Western North America Power System (wNAPS) and compared with a model-based approach for coordination of damping controllers. These are the first results of an extensive research related to coordination of DC-EIR, showing good adaptability and performance to different fault locations across the grid. 

Artificial intelligence-based prostate cancer (PCa) detection models have been widely explored to assist clinical diagnosis. However, these trained models may generate erroneous results specifically on datasets that are not within training distribution. In this paper, we propose an approach to tackle this so-called out-of-distribution (OOD) data problem. Specifically, we devise an end-to-end unsupervised framework to estimate uncertainty values for cases analyzed by a previously trained PCa detection model. Our PCa detection model takes the inputs of bpMRI scans and through our proposed approach we identify OOD cases that are likely to generate degraded performance due to the data distribution shifts. The proposed OOD framework consists of two parts. First, an autoencoder-based reconstruction network is proposed, which learns discrete latent representations of in-distribution data. Second, the uncertainty is computed using perceptual loss that measures the distance between original and reconstructed images in the feature space of a pre-trained PCa detection network. The effectiveness of the proposed framework is evaluated on seven independent data collections with a total of 1,432 cases. The performance of pre-trained PCa detection model is significantly improved by excluding cases with high uncertainty. 

3D CT point clouds reconstructed from the original CT images are naturally represented in real-world coordinates. Compared with CT images, 3D CT point clouds contain invariant geometric features with irregular spatial distributions from multiple viewpoints. This paper rethinks pulmonary nodule detection in CT point cloud representations. We first extract the multi-view features from a sparse convolutional (SparseConv) encoder by rotating the point clouds with different angles in the world coordinate. Then, to simultaneously learn the discriminative and robust spatial features from various viewpoints, a nodule proposal optimization schema is proposed to obtain coarse nodule regions by aggregating consistent nodule proposals prediction from multi-view features. Last, the multi-level features and semantic segmentation features extracted from a SparseConv decoder are concatenated with multi-view features for final nodule region regression. Experiments on the benchmark dataset (LUNA16) demonstrate the feasibility of applying CT point clouds in lung nodule detection task. Furthermore, we observe that by combining multi-view predictions, the performance of the proposed framework is greatly improved compared to single-view, while the interior texture features of nodules from images are more suitable for detecting nodules in small sizes. 

The early detection of where and when fatal infectious diseases outbreak is of critical importance to the public health. To effectively detect, analyze and then intervene the spread of diseases, people's health status along with their location information should be timely collected. However, the conventional practices are via surveys or field health workers, which are highly costly and pose serious privacy threats to participants. In this paper, we for the first time propose to exploit the ubiquitous cloud services to collect users' multi-dimensional data in a secure and privacy-preserving manner and to enable the analysis of infectious disease. Specifically, we target at the spatial clustering analysis using Kulldorf scan statistic and propose a key-oblivious inner product encryption (KOIPE) mechanism to ensure that the untrusted entity only obtains the statistic instead of individual's data. Furthermore, we design an anonymous and sybil-resilient approach to protect the data collection process from double registration attacks and meanwhile preserve participant's privacy against untrusted cloud servers. A rigorous and comprehensive security analysis is given to validate our design, and we also conduct extensive simulations based on real-life datasets to demonstrate the performance of our scheme in terms of communication and computing overhead. 

We report the first measurement of coherent elastic neutrino-nucleus scattering (CEvNS) on argon using a liquid argon detector at the Oak Ridge National Laboratory Spallation Neutron Source. Two independent analyses prefer CEvNS over the background-only null hypothesis with greater than 3σ significance. The measured cross section, averaged over the incident neutrino flux, is (2.2±0.7)×10−39  cm2—consistent with the standard model prediction. The neutron-number dependence of this result, together with that from our previous measurement on CsI, confirms the existence of the CEvNS process and provides improved constraints on nonstandard neutrino interactions.