Multiple Tiles Research Articles

DSM Reconstruction from Uncalibrated Multi-View Satellite Stereo Images by RPC Estimation and Integration

In this paper, we propose a 3D Digital Surface Model (DSM) reconstruction method from uncalibrated Multi-view Satellite Stereo (MVSS) images, where Rational Polynomial Coefficient (RPC) sensor parameters are not available. While recent investigations have introduced several techniques to reconstruct high-precision and high-density DSMs from MVSS images, they inherently depend on the use of geo-corrected RPC sensor parameters. However, RPC parameters from satellite sensors are subject to being erroneous due to inaccurate sensor data. In addition, due to the increasing data availability from the internet, uncalibrated satellite images can be easily obtained without RPC parameters. This study proposes a novel method to reconstruct a 3D DSM from uncalibrated MVSS images by estimating and integrating RPC parameters. To do this, we first employ a structure from motion (SfM) and 3D homography-based geo-referencing method to reconstruct an initial DSM. Second, we sample 3D points from the initial DSM as references and reproject them to the 2D image space to determine 3D–2D correspondences. Using the correspondences, we directly calculate all RPC parameters. To overcome the memory shortage problem while running the large size of satellite images, we also propose an RPC integration method. Image space is partitioned to multiple tiles, and RPC estimation is performed independently in each tile. Then, all tiles’ RPCs are integrated into the final RPC to represent the geometry of the whole image space. Finally, the integrated RPC is used to run a true MVSS pipeline to obtain the 3D DSM. The experimental results show that the proposed method can achieve 1.455 m Mean Absolute Error (MAE) in the height map reconstruction from multi-view satellite benchmark datasets. We also show that the proposed method can be used to reconstruct a geo-referenced 3D DSM from uncalibrated and freely available Google Earth imagery.

CFD studies into frictional effects of ship coatings to enhance condition monitoring & increase efficiency

Abstract Managing biofouling on ship hulls is essential for reducing CO2 emissions in the shipping industry. Biofouling, characterized by the accumulation of organisms on hulls, escalates surface roughness and hydrodynamic drag, subsequently increasing fuel consumption and CO2 emissions. In response, AkzoNobel, Royal Philips, and Damen Shipyards are developing a fouling prevention technology that utilizes Ultra Violet Light Emitting Diodes (UV-LEDs). The use of an array of multiple UV-LED tiles may lead to non-uniformities from gaps between tiles, potentially increasing frictional resistance due to variations in thickness and surface roughness at the tile joints. This study examines the impact of these non-uniformities on skin friction through simulations using the commercial Computational Fluid Dynamics (CFD) software, Star-CCM+. The analysis highlights effective jointing methods that minimize drag forces by optimizing gap-filling techniques, thus comparing favourably to traditional coatings. The findings contribute significantly to advancing a UV-C hull fouling prevention system, promising reduced environmental impacts and enhanced operational efficiency in the shipping industry.

Using YOLOv5, SAHI, and GIS with Drone Mapping to Detect Giant Clams on the Great Barrier Reef

Despite the ecological importance of giant clams (Tridacninae), their effective management and conservation is challenging due to their widespread distribution and labour-intensive monitoring methods. In this study, we present an alternative approach to detecting and mapping clam density at Pioneer Bay on Goolboddi (Orpheus) Island on the Great Barrier Reef using drone data with a combination of deep learning tools and a geographic information system (GIS). We trained and evaluated 11 models using YOLOv5 (You Only Look Once, version 5) with varying numbers of input image tiles and augmentations (mean average precision—mAP: 63–83%). We incorporated the Slicing Aided Hyper Inference (SAHI) library to detect clams across orthomosaics, eliminating duplicate counts of clams straddling multiple tiles, and further, applied our models in three other geographic locations on the Great Barrier Reef, demonstrating transferability. Finally, by linking detections with their original geographic coordinates, we illustrate the workflow required to quantify animal densities, mapping up to seven clams per square meter in Pioneer Bay. Our workflow brings together several otherwise disparate steps to create an end-to-end approach for detecting and mapping animals with aerial drones. This provides ecologists and conservationists with actionable and clear quantitative and visual insights from drone mapping data.

Spatial frequency decomposition with bandpass filters for multiscale analyses and functional correlations

To address the essential problem in surface metrology of establishing functional correlations spatial, frequencies in topographic measurements are progressively decomposed into a large number of narrow bands. Bandpass filters and commercially available software are used. These bands can be analyzed with conventional surface texture parameters, like average roughness, Sa, or other parameters, for detailed, multiscale topographic characterizations. Earlier kinds of multiscale characterization, like relative area, required specialized software performing multiple triangular tiling exercises. Multiscale regression analyses can test strengths of functional correlations over a range of scales. Here, friction coefficients are regressed against standard surface texture parameters over the range of scales available in a measurement. Correlation strengths trend with the scales of the bandpass filters. Using bandpass frequency, i.e., wavelength or scale, decompositions, the R2 at 25 μm, exceeds 0.9 for Sa compared with an R2 of only 0.2 using the broader band of conventional roughness filtering. These improved, scale-specific functional correlations can facilitate scientific understandings and specifications of topographies in product and process design and in designs of quality assurance systems.

Fast imaging of millimeter-scale areas with beam deflection transmission electron microscopy

Serial section transmission electron microscopy (TEM) has proven to be one of the leading methods for millimeter-scale 3D imaging of brain tissues at nanoscale resolution. It is important to further improve imaging efficiency to acquire larger and more brain volumes. We report here a threefold increase in the speed of TEM by using a beam deflecting mechanism to enable highly efficient acquisition of multiple image tiles (nine) for each motion of the mechanical stage. For millimeter-scale areas, the duty cycle of imaging doubles to more than 30%, yielding a net average imaging rate of 0.3 gigapixels per second. If fully utilized, an array of four beam deflection TEMs should be capable of imaging a dataset of cubic millimeter scale in five weeks.

On Computing Makespan-Optimal Solutions for Generalized Sliding-Tile Puzzles

In the 15-puzzle game, 15 labeled square tiles are reconfigured on a 4 × 4 board through an escort, wherein each (time) step, a single tile neighboring it may slide into it, leaving the space previously occupied by the tile as the new escort. We study a generalized sliding-tile puzzle (GSTP) in which (1) there are 1+ escorts and (2) multiple tiles can move synchronously in a single time step. Compared with popular discrete multi-agent/robot motion models, GSTP provides a more accurate model for a broad array of high-utility applications, including warehouse automation and autonomous garage parking, but is less studied due to the more involved tile interactions. In this work, we analyze optimal GSTP solution structures, establishing that computing makespan optimal solutions for GSTP is NP-complete and developing polynomial time algorithms yielding makespans approximating the minimum with expected/high probability constant factors, assuming randomized start and goal configurations.

Entropy and chirality in sphinx tilings

As a toy model of chiral interactions in crowded spaces, we consider sphinx tilings in finite regions of the triangular lattice. The sphinx tiles, hexiamonds composed of six equilateral triangles in the shape of a stylized sphinx, come in left and right enantiomorphs. Regions scaled up from the unit sphinx by an integer factor () require tiles of both chiral forms to produce tilings, including crystalline, quasicrystalline, and fully disordered tilings. For frames up to order 13, we describe methods that permit exact enumeration and computation of partition functions using , , and algorithms. For larger frames, we introduce a Monte Carlo method to sample typical tilings. The key to the latter is the identification of fundamental shapes () that admit multiple tilings and which allow a rejection-free MC simulation. Published by the American Physical Society 2024

Radial Magnetic Levitation and Its Application to Density Measurement, Separation, and Detection of Microplastics.

This work describes the development of radial magnetic levitation (MagLev) using two radially magnetized ring magnets to solve the problem of limited operational spaces in standard MagLev and the major shortcoming of a short working distance in axial MagLev. Interestingly and importantly, we demonstrate that for the same magnet size, this new configuration of MagLev doubles the working distance over the axial MagLev without significantly sacrificing the density measurement range, whether for linear or nonlinear analysis. Meanwhile, we develop a magnetic assembly method to fabricate the magnets for the radial MagLev, where multiple magnetic tiles with single-direction magnetization are used as assembly elements. On this basis, we experimentally demonstrate that the radial MagLev has good applicability in density-based measurement, separation, and detection and show its advantages in improving separation performance compared with the axial MagLev. The open structure of two-ring magnets and good levitation characteristics make the radial MagLev have great application potential, and the performance improvement brought by adjusting the magnetization direction of magnets provides a new perspective for the magnet design in the field of MagLev.

Abstract 5408: AI powered-platform to predict gene modifications from prostate and breast cancer whole slide images

Abstract There is little published evidence of predicting cancer genotypes directly from tissue histology, especially for breast and prostate cancers. Artificial intelligence (AI) enables discriminating and extracting morphological features reflective of the underlying genomic alterations at visual and subvisual levels. We have built a morphology-based and AI-powered platform for cancer genotyping, risk stratification and outcome prediction that addresses the needs for treatment decision-making in a cost effective and timely fashion. A cohort of 390 prostate and 742 invasive breast cancer patients with known molecular status of key genes, such as TP53, PIK3CA, MYC, ERBB2, TMPRSS2-ERG and PTEN from The Cancer Genome Atlas were included in this study. Hematoxylin and eosin (H&E) stained whole slide images (WSI) of the cancer tissue sections were available at 20x or 40x. The WSI from the two cancer cohorts were split 2:1 into a training and test dataset, respectively. Our platform involved two different deep Convolutional Neural Network (DCNN) architectures. The platform first divided each WSI into multiple tiles. Each tile was then analyzed using a DCNN that graded the tile and generated a high dimensional vector to provide a mathematical representation of the morphology. The combination of high dimensional vectors across the WSI was then fed into a second DCNN that generated a morphological score, which predicted whether the gene under consideration was wild type or modified. Our platform has achieved 70 - 80% accuracy as defined by the Area under the Curve for the receiver operating characteristics curves for the genetic markers on the test cohorts (Table 1). Our platform can predict genotypes/molecular alterations directly from H&E stained WSI with high accuracy. This technology presents a novel, practical and cost-effective approach for cancer molecular classification and risk stratification, enabling timely and optimal treatment decision-making for positive clinical outcomes. Genotype Prediction for Breast Cancer (BrCa) and Prostate Cancer (PCa) from WSI Cohort Molecular Biomarker Training Dataset (n) Test Dataset (n) Test ROC AUC Score (%) Gene Modification/Loss Intact/Wild type Gene Modification/Loss Intact/Wild type BrCa TP53 196 302 96 149 80 PIK3CA 153 344 75 170 74 MYC 68 414 33 204 77 ERBB2 66 416 32 205 78 PCa TMPRSS2-ERG 104 157 51 78 70 PTEN 57 201 28 99 73 Citation Format: Wei Huang, Parag Jain, Chensu Xie, Hassan Muhammad, Hirak Basu, George Wilding, Rajat Roy. AI powered-platform to predict gene modifications from prostate and breast cancer whole slide images. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5408.

Dynamic Region of Interest Selection in Remote Photoplethysmography: Proof-of-Concept Study.

Remote photoplethysmography (rPPG) can record vital signs (VSs) by detecting subtle changes in the light reflected from the skin. Lifelight (Xim Ltd) is a novel software being developed as a medical device for the contactless measurement of VSs using rPPG via integral cameras on smart devices. Research to date has focused on extracting the pulsatile VS from the raw signal, which can be influenced by factors such as ambient light, skin thickness, facial movements, and skin tone. This preliminary proof-of-concept study outlines a dynamic approach to rPPG signal processing wherein green channel signals from the most relevant areas of the face (the midface, comprising the cheeks, nose, and top of the lip) are optimized for each subject using tiling and aggregation (T&A) algorithms. High-resolution 60-second videos were recorded during the VISION-MD study. The midface was divided into 62 tiles of 20×20 pixels, and the signals from multiple tiles were evaluated using bespoke algorithms through weighting according to signal-to-noise ratio in the frequency domain (SNR-F) score or segmentation. Midface signals before and after T&A were categorized by a trained observer blinded to the data processing as 0 (high quality, suitable for algorithm training), 1 (suitable for algorithm testing), or 2 (inadequate quality). On secondary analysis, observer categories were compared for signals predicted to improve categories following T&A based on the SNR-F score. Observer ratings and SNR-F scores were also compared before and after T&A for Fitzpatrick skin tones 5 and 6, wherein rPPG is hampered by light absorption by melanin. The analysis used 4310 videos recorded from 1315 participants. Category 2 and 1 signals had lower mean SNR-F scores than category 0 signals. T&A improved the mean SNR-F score using all algorithms. Depending on the algorithm, 18% (763/4212) to 31% (1306/4212) of signals improved by at least one category, with up to 10% (438/4212) improving into category 0, and 67% (2834/4212) to 79% (3337/4212) remaining in the same category. Importantly, 9% (396/4212) to 21% (875/4212) improved from category 2 (not usable) into category 1. All algorithms showed improvements. No more than 3% (137/4212) of signals were assigned to a lower-quality category following T&A. On secondary analysis, 62% of signals (32/52) were recategorized, as predicted from the SNR-F score. T&A improved SNR-F scores in darker skin tones; 41% of signals (151/369) improved from category 2 to 1 and 12% (44/369) from category 1 to 0. The T&A approach to dynamic region of interest selection improved signal quality, including in dark skin tones. The method was verified by comparison with a trained observer's rating. T&A could overcome factors that compromise whole-face rPPG. This method's performance in estimating VS is currently being assessed. ClinicalTrials.gov NCT04763746; https://clinicaltrials.gov/ct2/show/NCT04763746.

TVSR‐OR: Tile‐based 360‐degree video streaming over real time streaming protocol with optimized read

AbstractAmong video application use cases and scenarios, 360‐degree video applications gradually become significant. The state‐of‐the‐art solution for 360‐degree video streaming is field of view (FOV) transmission, whose quality of experience highly depends on system processing speed and network latency. The extant 360‐degree video FOV transmission profiles transmit multiple video tile streams to the client over either Hypertext Transfer Protocol (HTTP) or Real Time Streaming Protocol (RTSP). The client needs to merge all different tile streams after receiving them, which may cause a data synchronization problem of waiting for each tile to sync in the presents of packet losses. Therefore, we propose a system to select and merge tile streams on the Content Delivery Network (CDN) server and transmit the merged stream to the client. During the streaming service, the client can send a viewport switch request through RTSP signaling, and the server will subsequently deliver the merged video stream of the new viewport to the client after receiving the request. Also, to reduce disk overhead from parallel reads, we optimize the system with a file prefetch strategy to reduce unorganized, random, and parallel reads into serialized reads of large data trunks. We evaluate its performance through user request simulation experiments. Compared with the extant transmission solution, our system provides slightly lower video transmission latency, lower central processing unit (CPU) usage, and better disk performance under heavy service load.

Accelerating Deep Convolutional Neural Networks Using Number Theoretic Transform

Modern deep convolutional neural networks (CNNs) suffer from high computational complexity due to excessive convolution operations. Recently, fast convolution algorithms such as fast Fourier transform (FFT) and Winograd transform have gained attention to address this problem. They reduce the number of multiplications required in the convolution operation by replacing it with element-wise multiplication in the transform domain. However, fast convolution-based CNN accelerators have three major concerns: expensive domain transform, large memory overhead, and limited flexibility in kernel size. In this paper, we present a novel CNN accelerator based on number theoretic transform (NTT), which overcomes the existing limitations. We propose the low-cost NTT and inverse-NTT converter that only use adders and shifters for on-chip domain transform, which solves the inflated bandwidth problem and enables more parallel computations in the accelerator. We also propose the accelerator architecture that includes multiple tile engines with the optimized data flow and mapping. Finally, we implement the proposed NTT-based CNN accelerator on the Xilinx Alveo U50 FPGA and evaluate it for popular deep CNN models. As a result, the proposed accelerator achieves 2859.5, 990.3, and 805.6 GOPS throughput for VGG-16, GoogLeNet, and Darknet-19, respectively. It outperforms the existing fast convolution-based CNN accelerators up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$9.6\times $ </tex-math></inline-formula> .

MSS12 Presentation Time: 4:40 PM: A Proposed Formalism for Postoperative Dosimetry Evaluation for Permanently Implanted Collagen Tile Embedded Cs-131 Brachytherapy in the Brain

<h3>Purpose</h3> To propose a standardized formalism for delineating target volume contours and evaluating dose on postoperative imaging for patients receiving permanent low dose rate (LDR) brachytherapy (BT) in the brain. The BT applicator considered for this work consists of a 20 × 20 × 4 mm collagen tile embedded with four 3.5 U Cs-131 seeds with a 10 mm inter-seed spacing (GammaTile®, GT Medical Technologies, USA). Following maximum safe resection, multiple tiles are used to line the cavity and deliver a nominal dose of 60 Gy to a depth of 5 mm from the operative bed surface. The tiles are designed to limit the maximum dose to brain tissue via a 3 mm seed offset from brain surface that functions as an implanted internal compensator (Fig 1A). <h3>Materials and Methods</h3> Collagen tiles are placed by neurosurgeons under surgical visualization on resection the surfaces suspected to harbor subclinical disease; in this setting the Clinical Target Volume (CTV) is defined as the region of parenchyma adjacent to tile placement to a depth of 5 mm. As illustrated in Fig 1B-D, the algorithm proposed to generate a CTV is): 1) outline the brain, 2) outline all seeds using a Hounsfield Unit-based threshold tool, 3) expand the seed contour by 3 mm to approximate the tile/brain interface, 4) outline the cavity visualized on a registered MRI using the seeds+3 mm contour as a guide, 5) use a Boolean operation to obtain the intersection of the cavity contour expanded by 5 mm, the seed contour expanded by 0.94 mm, and the brain contour, and 5) subtract the cavity. Manual adjustments to ensure only areas with suspected subclinical disease may be required. Postoperative dose was calculated as per the AAPM TG43 3D formalism using the consensus dataset for the Cs-131 seed model utilized in the GammaTile device. Dose statistics pertaining to the CTV, including the mean isodose values covering 90% and 95% of the volume (D90% and D95%), the volume receiving at least 90% and 95% of the prescription 60 Gy (V90% and V95%), minimum dose, and maximum dose were tabulated for 11 patients using a commercial Treatment Planning System (MIM, MIM Software, USA). <h3>Results</h3> Volumes (cc) of cavity contours across all patients ranged from 3.3 to 49.9 and the CTVs ranged from 6.9 to 27.1. Mean D90% and D95% (Gy) to the CTV was 55.8 (45.5 - 73.8) and 49.4 (36.3 - 67.1), respectively. Mean V90% and V95% (% of total volume) to the CTV was 88.7 (77.6 - 99.8) and 85.5 (72.6 - 99.4), respectively. Mean minimum and maximum doses (Gy) to the CTV were 36.3 (19. - 64.7) and 426.3 Gy (265.6 - 573.9), respectively. Note, the maximum dose to the <i>cavity</i> is not accurately calculated by the TPS but is estimated to be in the range of 800 - 1500 Gy, indicating the tiles reduce maximum dose to brain by ∼28-53%. <h3>Conclusions</h3> A formalism to 1) generate a CTV representing brain containing subclinical disease following gross tumor resection and collagen tile BT implant and 2) obtain dose statistics is proposed. The contouring algorithm utilizes the surgical cavity (well-seen on MRI), LDR seeds (placed with direct surgical guidance, and well-seen on CT) and Boolean operations, resulting in a straightforward and reproducible CTV. The proposed dose statistics (D90%, D95%, V90%, V95%, min, and max) proved to be useful metrics for dose review. The standardized formalism for contouring CTV and its use in dose evaluation will be further tested to determine its utility in minimizing interobserver variability. Once fully validated, this formalism will be used to generate recommendations for quantitative evaluation GammaTile implant dosimetry.

Characterization of the three-dimensional multiple translative tiles

This paper proves the following statement which answers questions of Gravin, Robins and Shiryaev: If a convex body can form a two, three or fourfold translative tiling inE3, it must be a lattice tile (parallelohedron). In other words, it must be a parallelotope, a hexagonal prism, a rhombic dodecahedron, an elongated dodecahedron, or a truncated octahedron. As a comparison, it is known that there exist non-parallelohedral fivefold translative tiles in every dimension.

Cross-User Similarities in Viewing Behavior for 360° Video and Caching Implications

The demand and usage of 360° video services are expected to increase. However, despite these services being highly bandwidth intensive, not much is known about the potential value that basic bandwidth saving techniques such as server or edge-network on-demand caching (e.g., in a CDN) could have when used for delivery of such services. This problem is both important and complicated as client-side solutions have been developed that split the full 360° view into multiple tiles, and adapt the quality of the downloaded tiles based on the user’s expected viewing direction and bandwidth conditions. This article presents new trace-based analysis methods that incorporate users’ viewports (the area of the full 360° view the user actually sees), a first characterization of the cross-user similarities of the users’ viewports, and a trace-based analysis of the potential bandwidth savings that caching-based techniques may offer under different conditions. Our analysis takes into account differences in the time granularity over which viewport overlaps can be beneficial for resource saving techniques, compares and contrasts differences between video categories, and accounts for uncertainties in the network conditions and the prediction of the future viewing direction when prefetching. The results provide substantial insight into the conditions under which overlap can be considerable and caching effective, and inform the design of new caching system policies tailored for 360° video.

User Gaze-Driven Adaptation of Omnidirectional Video Delivery Using Spatial Tiling and Scalable Video Encoding

Omnidirectional video is becoming increasingly popular among viewers, but its delivery requires considerable amount of network bandwidth. Today’s streaming services are transmitting the full spatial angle of omnidirectional videos, although most of the transmitted content is not utilized. Due to both limited bandwidth availability and its dynamic fluctuations, adaptive delivery solutions play a key role in supporting high user quality streaming of omnidirectional videos. This paper describes research which extends the MPEG-DASH Spatial Relationship Description by adding scalable video encoding to spatial tiling. It proposes a novel tile-layering based gaze adaptation algorithm for omnidirectional video delivery and employs it in conjunction with multiple tiling schemes. The benefits of the proposed algorithm with diverse tiling schemes are evaluated objectively in terms of bandwidth savings and adaptation latency. The results show a reduction of network bandwidth requirements to about 30% of the original bandwidth value with a low processing latency of 70.87ms.

Fast and accurate multi-class geospatial object detection with large-size remote sensing imagery using CNN and Truncated NMS

Multi-class geospatial object detection with remote sensing imagery has broad prospects in urban planning, natural disaster warning, industrial production, military surveillance and other applications. Accuracy and efficiency are two common measures for evaluating object detection models, and it is often difficult to achieve both at the same time. Developing a practical remote sensing object detection algorithm that balances the accuracy and efficiency is thus a big challenge in the Earth observation community. Here, we propose a comprehensive high-speed multi-class remote sensing object detection method. Firstly, we obtain a multi-volume YOLO (You Only Look Once) v4 model for balancing speed and accuracy, based on a pruning strategy of the convolutional neural network (CNN) and the one-stage object detection network YOLO v4. Moreover, we apply the Manhattan-Distance Intersection of Union (MIOU) loss function to the multi-volume YOLO v4 to further improve the accuracy without additional computational burden.Secondly, mainly due to computing limitations, a remote sensing image that is large-size relative to a natural image must first be divided into multiple smaller tiles, which are then detected separately, and finally, the detection results are spliced back to match the original image. In the process of remote sensing image slicing, a large number of truncated objects appear at the edge of tiles, which will produce a large number of false results in the subsequent detection links. To solve this problem, we propose a Truncated Non-Maximum Suppression (NMS) algorithm to filter out repeated and false detection boxes from truncated targets in the spliced detection results. We compare the proposed algorithm with the state-of-the-art methods on the Dataset for Object deTection in Aerial images (DOTA) and DOTA v2. Quantitative evaluations show that mAP and FPS reach 77.3 and 35 on DOTA, and 61.0 and 74 on DOTA v2. Overall, our method reaches the optimal balance between efficiency and accuracy, and realizes the high-speed remote sensing object detection.

Optimal Capacity-Driven Design of Aperiodic Clustered Phased Arrays for Multi-User MIMO Communication Systems

The optimal design of aperiodic/irregular clustered phased arrays (PAs) for base stations (BSs) in multi-user (UE) multiple-input–multiple-output (MU-MIMO) communication systems is addressed. This article proposes an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ad hoc</i> synthesis method aimed at maximizing the UEs traffic capacity within the cell served by the BS while guaranteeing a sufficient level of the signal at the terminals. Toward this end, the search of the optimal aperiodic clustering is carried out through a customized tiling technique able to consider both single and multiple tile shapes and assure the complete coverage of the antenna aperture for the maximization of the directivity. Representative results, from a wide set of numerical examples concerned with realistic antenna models and benchmark Third Generation Partnership Project (3GPP) scenarios, are reported to assess the advantages of the irregular array architectures in comparison with regular/periodic layouts proposed by the standard development organizations as well.

Artificial Intelligence Improves Pathologist Agreement for Fibrosis Scores in Nonalcoholic Steatohepatitis Patients

Artificial Intelligence Improves Pathologist Agreement for Fibrosis Scores in Nonalcoholic Steatohepatitis Patients

DISSEC: A distributed deep neural network inference scheduling strategy for edge clusters

New applications such as intelligent manufacturing, autonomous vehicles and smart cities drive large-scale deep learning models deployed in the Internet of Things (IoT) edge environments. However, deep learning models require substantial computations, storage and communication resources to run. It is generally difficult to deploy and execute a complete deep neural network (DNN) on a resource-constrained edge device. One possible solution is to slice the DNN into multiple tiles distributed to different edge devices, which can reduce the number of computations and quantity of data on each edge device. In this paper, we propose DISSEC, a distributed scheduling strategy for DNN inference on IoT edge clusters. DISSEC leverages spatial partitioning techniques through fusing the convolutional layers and dividing them into multiple partitions that can be executed independently, and proposes a method to express the dependencies between partitions. It further proposes a search algorithm based on heuristics to produce a distributed parallel strategy with the best overall inference execution latency. The evaluation shows that our strategy can fully utilize the edge device resources by cooperating with multiple edge devices to perform partitioning tasks in parallel. Furthermore, compared to the existing work scheduling strategy, our strategy reduces communication overhead by 20% and overall execution latency by 9% under different partitioning granularities and numbers of edge devices.