Software Library Research Articles

HRA-Secure Homomorphic Lattice-Based Proxy Re-Encryption with Tight Security

We construct an efficient proxy re-encryption (PRE) scheme secure against honest re-encryption attacks (HRA-secure) with precise concrete security estimates. To get these precise concrete security estimates, we introduce the tight, fine-grained noise-flooding techniques of Li et al. (CRYPTO'22) to RLWE-based (homomorphic) PRE schemes, as well as a mixed statistical-computational security definition to HRA security analysis. Our solution also supports homomorphic operations on the ciphertexts. Such homomorphism allows for advanced applications, e.g., encrypted computation of network statistics across networks, and unlimited hops in the case of full homomorphism, i.e., when bootstrapping is available. We implement our PRE scheme in the OpenFHE software library and apply it to a problem of secure multi-hop data distribution in the context of 5G virtual network slices. We also experimentally evaluate the performance of our scheme, demonstrating that the implementation is practical. Moreover, we compare our PRE method with other lattice-based PRE schemes and approaches targeting HRA security. These achieve HRA security, but not in a tight, practical scheme such as our work. Further, we present an attack on the PRE scheme proposed in Davidson et al.'s (ACISP'19), which was claimed to achieve HRA security without noise flooding, i.e., without adding large noise.

Glymphatic System in Preterm Neonates: Developmental Insights Following Birth Asphyxia.

Birth asphyxia (BA) and germinal matrix hemorrhage-intraventricular hemorrhage (GMH-IVH) are common clinical events in preterm neonates. However, their effects on the glymphatic system (GS) development in preterm neonates remain arcane. To evaluate the developmental trajectory of the GS, and to investigate the effects of BA and GMH-IVH on GS function in preterm neonates. Prospective. Two independent datasets, prospectively acquired internal dataset (including 99 preterm neonates, 40 female, mean [standard deviation] gestational age (GA) at birth, 29.95 [2.63] weeks) and the developing Human Connectome Project (dHCP) dataset (including 81 preterm neonates, 29 female, median [interquartile range] GA at birth, 32.71 [4.28] weeks). 3.0 T MRI and diffusion-weighted spin-echo planar imaging sequence. The diffusion-weighted images were preprocessed in volumetric space using the FMRIB Software Library and diffusion along the perivascular space (DTI-ALPS) index was accessed to evaluate GS function. Two sample t tests, one-way analysis of variance followed by least-significant difference (LSD) post hoc analysis, chi-squared tests, and Pearson's correlation analysis. Significance level: P < 0.05. In prospectively acquired internal dataset, preterm neonates with BA exhibited a significant lower DTI-ALPS index than those without BA (0.98 ± 0.08 vs. 1.08 ± 0.07, T = -5.89); however, GMH-IVH did not exert significant influences on the DTI-ALPS index (P = 0.83 and 0.27). The DTI-ALPS index increased significantly at postmenstrual age ranging from 25 to 34 weeks (r = 0.38) and then plateaued after 34 weeks (P = 0.35), which we also observed in the dHCP dataset. BA rather than GMH-IVH serves as the major influencing factor in the development of GS in preterm neonates. Moreover, as GS development follows a nonlinear trajectory, we recommend close monitoring of GS development in preterm neonates with a GA less than 34 weeks. 2 TECHNICAL EFFICACY: Stage 2.

Comparison of software implementations of the Bundle Adjustment algorithm for the SLAM problem

This paper investigates the problem of Simultaneous Localization and Mapping (SLAM) with a focus on the optimization stage of Bundle Adjustment (BA). This stage is crucial for accurately reconstructing the 3D structure of a scene and estimating camera parameters, which directly impact subsequent computer vision tasks such as tracking, depth map generation, and reconstruction. Modern software libraries implementing the BA algorithm are reviewed, including Ceres Solver, GTSAM, and SymForce. An experimental comparison of these libraries is conducted based on performance criteria such as execution speed, number of iterations, Absolute Pose Error (APE), and Cost Change. The BAL (Bundle Adjustment in the Large) dataset and three specific datasets—Ladybug, Trafalgar Square, and Dubrovnik—were used for testing. A software solution was developed to evaluate the performance of these libraries, incorporating steps such as normalization, adding random noise, and executing optimization algorithms. The visualization of solutions was performed using the Evo library. Experimental results show that SymForce provides the best balance between speed, accuracy, and convergence. It achieved the shortest execution time for the Ladybug and Trafalgar Square datasets and exhibited the lowest Absolute Pose Error across all cases. Ceres Solver proved competitive, particularly for the Dubrovnik dataset, where it outperformed SymForce in terms of speed. In contrast, GTSAM demonstrated the worst performance across all criteria, exhibiting the highest absolute errors and the longest computation times. Based on these findings, SymForce is the most suitable tool for SLAM and nonlinear optimization tasks where both speed and accuracy are critical. Ceres Solver can be effective in specific scenarios requiring maximum computational speed. Meanwhile, GTSAM lags behind its competitors, limiting its applicability in high-performance systems. These conclusions will aid in selecting the appropriate optimization libraries for practical applications in robotics, autonomous vehicles, and 3D mapping.

TriMe++: Multi-threaded triangular meshing in two dimensions

We present TriMe++, a multi-threaded software library designed for generating two-dimensional meshes for intricate geometric shapes using the Delaunay triangulation. Multi-threaded parallel computing is implemented throughout the meshing procedure, making it suitable for fast generation of large-scale meshes. Three iterative meshing algorithms are implemented: the DistMesh algorithm, the centroidal Voronoi diagram meshing, and a hybrid of the two. We compare the performance of the three meshing methods in TriMe++, and show that the hybrid method retains the advantages of the other two. The software library achieves significant parallel speedup when generating a large mesh with 106 points. TriMe++ can handle complicated geometries and generates adaptive meshes of high quality. Program summaryProgram title:TriMe++CPC Library link to program files:https://doi.org/10.17632/jxcsxtywtw.1Developer's repository link:https://github.com/jiayinlu19960224/TriMeLicensing provisions: BSD 3-clauseProgramming language: C++External routines/libraries: OpenMP, multi-threaded Voro++Nature of problem: Multi-threaded geometry meshing in two dimension using the Delaunay triangulationSolution method: The TriMe++ library is built around several C++ classes that follows a structured meshing pipeline. During initialization, the shape_2d class reads the geometry input and generates a signed distance field using a grid-based data structure to represent the shape. The sizing_2d class subsequently produces adaptive element sizing and density fields for the mesh. It uses an adaptive quad-tree data structure, enabling efficient refinement of sizing and density values in areas with complex geometries. In the meshing procedure, the parallel_meshing_2d class iteratively improves point positions in the mesh. In each meshing iteration, the multi-threaded Voro++ library generates the Delaunay triangulation of the points. Users can select from three meshing algorithms, the DistMesh algorithm in the mesh_alg_2d_dm class, the centroidal Voronoi diagram meshing algorithm in the mesh_alg_2d_cvd class, and a hybrid method of the two in the mesh_alg_2d_hybrid class Throughout this meshing workflow, we use OpenMP for multi-threaded parallel computations.

State-of-the-Art Computer Vision and ML Operations: An Integrated Analysis of Software and Hardware Architectures

This comprehensive article examines the current state of computer vision and machine learning operations across software frameworks and hardware platforms. It explores the evolving landscape of CV/ML deployments, analyzing the integration challenges and optimization strategies in modern systems. This article evaluates prominent software libraries and frameworks with primary usecases. It encompasses hardware acceleration techniques, dedicated neural processors, and vision processing units, offering insights into their performance characteristics and power efficiency trade-offs. It highlights emerging trends in hardware-software co-design, standardization efforts, and future directions in the field. This article contributes to the understanding of optimal CV/ML deployment strategies while addressing the growing demands of edge computing and real-time processing requirements.

Open-Source DNA-Encoded Library Package for Design, Decoding and Analysis: DELi.

DNA-encoded library (DEL) technology has become a powerful tool in modern drug discovery. Fully harnessing its potential requires the use of advanced computational methodologies, which are often available only through proprietary software. This limitation restricts flexibility and accessibility for academic researchers and small biotech companies, hindering the growth of the technology. Here, we present DELi, an open-source DEL informatics platform designed for library design, NGS decoding and calling, and enrichment analysis. To showcase its capabilities, we used DELi to design an in-house custom library (UNC-DEL006), a benzimidazole-based DEL, and performed proof-of-concept selection experiments against Bromodomain-containing Protein 4 (BRD4). The DELi decoding and analysis modules identified top-performing compounds, leading to the off-DNA synthesis of UNC 002-080, which was confirmed as a nanomolar BRD4 binder via isothermal titration calorimetry (ITC). In contrast, a chemically similar compound not prioritized by DELi, UNC 002-083, showed no measurable binding. These results demonstrate DELi as an effective tool for DEL design and analysis. Further, its open-source nature will promote ongoing development and contributions from the DEL community to expand its applications and capabilities.

МОДУЛЬ АНАЛІЗУ ОПИТУВАНЬ ЗДОБУВАЧІВ МАЛОКОМПЛЕКТНИХ ГРУП У СИСТЕМІ KSU24

Feedback is one of the most effective tools for quality control in education, used by higher education institutions in Ukraine and around the world. It enables the collection of information regarding students' perception of educational materials and facilitates analysis of learning outcomes for a better understanding of the current effectiveness of a specific discipline or educational component within an academic program, with the goal of improving the quality of the discipline and, thus, the overall academic program. Feedback can relate to an entire discipline, a specific semester, the module in which the discipline is taught, or even an individual class session. A wide range of mathematical methods and models, as well as ready-made software libraries, allow for qualitative analysis of feedback results, with most requiring an adequate amount of information for processing. However, in Ukrainian higher education institutions, including Kherson State University, there are academic groups in certain specialties where the number of students is insufficient for applying most analytical methods, especially at the master’s level or higher, or in elective disciplines. To obtain information on learning effectiveness in these groups, equivalent to that in groups with a sufficient number of students, specialized analysis methods are needed. The article describes methods for analyzing feedback results in the form of surveys within the KSU24 system, allowing for consideration of relatively small student groups. It discusses approaches for the software implementation of relevant algorithmic methods and necessary architectural solutions for building a survey module within the KSU24 system. A key functional requirement of this module is ensuring anonymity in the process, as integrity among participants is critically important for obtaining high-quality feedback, especially when processing a limited number of results.

Awareness and Skillfulness in Using Arabic Language Instructional Software Among Senior Secondary Schools in Kwara State

The study examined awareness and skillfulness in using Arabic Instructional software among senior secondary schools in Kwara state. A descriptive survey research design was used. Arabic language teachers were the respondents. There were 146 Arabic language teachers in Kwara State (Kwara State Teaching Service Commission, 2023). A sample of 108 Arabic language teachers was selected using the Kredjcie and Morgan sample size determination table. The research instrument is "Awareness and Skillfulness in Using Instructional Software Questionnaire (ASUISQ)". Findings showed that there is low awareness of Arabic language Instructional software among the majority of Arabic teachers in senior secondary schools in Kwara state is relatively low. The study indicates low skillfulness in using Arabic language Instructional software among Arabic teachers in senior secondary schools in Kwara state. Based on the results, there is low awareness of Verbace Arabic-English Offline Dictionary, Best Arabic Calligraphy and Weyyak Arabic Library Software, SHAHID Arabic Video Streaming App and Smart 13rab Software as instructional software among Arabic teachers in senior secondary schools in Kwara state is relatively low. The study also concluded that there is low skillfulness of Verbace Arabic-English Offline Dictionary, best Arabic Calligraphy and Weyyak Arabic Library Software, SHAHID Arabic Video Streaming App and Smart 13rab Software as instructional software among Arabic teachers in senior secondary schools in Kwara State. The study also recommended that the Kwara State Ministry of Education make adequate budgetary allocations for in-service training, particularly on modern instructional software.

Randomized tensor decomposition using parallel reconfigurable systems

Tensor decomposition algorithms are essential for extracting meaningful latent variables and uncovering hidden structures in real-world data tensors. Unlike conventional deterministic tensor decomposition algorithms, randomized methods offer higher efficiency by reducing memory requirements and computational complexity. This paper proposes an efficient hardware architecture for a randomized tensor decomposition implemented on a field-programmable gate array (FPGA) using high-level synthesis (HLS). The proposed architecture integrates random projection, power iteration, and subspace approximation via QR decomposition to achieve low-rank approximation of multidimensional datasets. The proposed architecture utilizes the capabilities of reconfigurable systems to accelerate tensor computation. It includes three central units: (1) tensor times matrix chain (TTMc), (2) tensor unfolding unit, and (3) QR decomposition unit to implement a three-stage algorithm. Experimental results demonstrate that our FPGA design achieves up to 14.56 times speedup compared to the well-implemented tensor decomposition using software library Tensor Toolbox on an Intel i7-9700 CPU. For a large input tensor of size 512×512×512, the proposed design achieves a 5.55 times speedup compared to an Nvidia Tesla T4 GPU. Furthermore, we utilize our hardware-based high-order singular value decomposition (HOSVD) accelerator for two real applications: background subtraction of dynamic video datasets and data compression. In both applications, our proposed design shows high efficiency regarding accuracy and computational time.

Selective State Space Models Outperform Transformers at Predicting RNA-Seq Read Coverage.

Transformers are the basis for many state-of-the-art machine learning tools, including those for predicting gene expression data from DNA sequence. The considerable time and cost of training transformer models has motivated development of alternative approaches inspired by ideas from the signal-processing literature, such as state-space models (Mamba), Fourier transforms (Hyena), and wavelet transforms (MultiResNet). To evaluate these methods as potential replacements (or complements) for attention, we developed a software library bilby, implemented using Python and Jax/Flax, providing convolutional, attention, bidirectional Hyena, bidirectional Mamba, and striped-architecture models for supervised multi-task learning in functional genomics. We report a comparison of these architectures, testing several hyperparameters and variations, and reporting performance statistics for the withheld test set as well as downstream SNP classifiers. Relative to models comprising convolution and attention layers (implemented in Python and TensorFlow via the Baskerville library used by the Borzoi software), models comprising convolutional, bidirectional Mamba, and (optionally) attention layers achieve small but consistent improvements in prediction accuracy, for roughly comparable training times and parameter counts, when averaged across all output tracks and data splits (a proportional increase of 3-4% in Pearson R, and 1-2% in r 2 , with the highest gains achieved when Mamba and attention layers were combined in a striped architecture). In contrast, Hyena (when reimplemented as described in the literature) was not competitive with attention-based models at these tasks, while MultiResNet proved too slow to be practical. The gains in prediction accuracy of the Mamba-based models do not yet translate to significantly improved performance on downstream SNP classification tasks: benchmarks using a GTEx eQTL dataset yield roughly similar results for Mamba- and attention-based classifiers, with attention marginally outperforming Mamba in one metric (a difference of +0.007 in area under ROC) and slightly underperforming by another metric (a difference of -0.006 in Spearman rank correlation). We argue that these results suggest selective state-space models (such as Mamba and Striped Mamba) warrant further exploration for functional genomics tasks. Our code and trained models are publicly available at https://github.com/ihh/bilby .

Implementation of Visual Odometry on Jetson Nano.

This paper presents the implementation of ORB-SLAM3 for visual odometry on a low-power ARM-based system, specifically the Jetson Nano, to track a robot's movement using RGB-D cameras. Key challenges addressed include the selection of compatible software libraries, camera calibration, and system optimization. The ORB-SLAM3 algorithm was adapted for the ARM architecture and tested using both the EuRoC dataset and real-world scenarios involving a mobile robot. The testing demonstrated that ORB-SLAM3 provides accurate localization, with errors in path estimation ranging from 3 to 11 cm when using the EuRoC dataset. Real-world tests on a mobile robot revealed discrepancies primarily due to encoder drift and environmental factors such as lighting and texture. The paper discusses strategies for mitigating these errors, including enhanced calibration and the potential use of encoder data for tracking when camera performance falters. Future improvements focus on refining the calibration process, adding trajectory correction mechanisms, and integrating visual odometry data more effectively into broader systems.

Applications of Open Source Software in Libraries

Applications of Open Source Software in Libraries

Targeted screening and profiling of massive components of colistimethate sodium by two-dimensional-liquid chromatography-mass spectrometry based on self-constructed compound database

Targeted screening and profiling of massive components of colistimethate sodium by two-dimensional-liquid chromatography-mass spectrometry based on self-constructed compound database

Deviant sound frequency and time stimuli in auditory oddball tasks reveal persistent aberrant brain activity in patients with psychosis and symptomatic remission.

Deviant sound frequency and time stimuli in auditory oddball tasks reveal persistent aberrant brain activity in patients with psychosis and symptomatic remission.

Brain MRI Detection of an Abnormal Peak Width of Skeletonized Mean Diffusivity in REM Sleep Behavior Disorder.

Peak width of skeletonized mean diffusivity (PSMD) is a novel marker of white matter damage, which may be related to small vessel disease. This study aimed to investigate the presence of white matter damage in patients with isolated rapid eye movement sleep behavior disorder (RBD) using PSMD. We enrolled patients with newly diagnosed isolated RBD confirmed by polysomnography and age- and sex-matched healthy controls. Diffusion tensor imaging (DTI) was conducted using a 3-Tesla MRI scanner. We measured the PSMD based on DTI in several steps, including preprocessing, skeletonization, application of a custom mask, and histogram analysis, using the Functional Magnetic Resonance Imaging of the Brain Software Library program. We compared the incidence of PSMD between patients with RBD and healthy controls and performed a correlation analysis between PSMD and clinical factors in patients with RBD. Thirty patients with isolated RBD and 41 healthy controls were enrolled. The PSMD was significantly higher in patients with RBD than that in the healthy controls (3.078vs. 2.746×10-4mm2/s, p=0.001). In addition, PSMD positively correlated with age in patients with RBD (r=0.477, p=0.007). However, PSMD was not associated with other clinical or polysomnographic factors. Patients with isolated RBD had a higher PSMD than healthy controls, indicating the evidence of white matter damage in patients with RBD. This finding highlights the potential of PSMD as a marker for detecting white matter damage, which may be related to small vessel diseases, in patients with sleep disorders.

Comparison of a Minimization and a Saddle Point Formulation of a Coupled Mechanics‐Diffusion Problem Implemented in deal.II

AbstractFor a problem in chemo‐mechanics, stated as a rate‐type variational principle, a minimization and a saddle point formulation are considered. Both formulations couple mechanical balance equations to mass diffusion and are intended to accurately model the diffusion‐induced swelling of hydrogels. The variational formulation is discretized by finite elements and implemented using the deal.II software library. The linearized systems in the incremental‐iterative solution procedure are solved using the unsymmetric multifrontal package (UMFPACK) sparse direct solver. The two formulations, employing three different, conforming finite element types, are considered for the case of mechanically‐induced diffusion and compared with respect to the evolution of the reaction force, mesh convergence, and the solution time. This initial step, comparing the performance of sparse direct solvers, sets the stage for future assessments involving domain decomposition, where sparse direct solvers are a computational kernel.

Research Directions in Software Supply Chain Security

Reusable software libraries, frameworks, and components, such as those provided by open-source ecosystems and third-party suppliers, accelerate digital innovation. However, recent years have shown almost exponential growth in attackers leveraging these software artifacts to launch software supply chain attacks. Past well-known software supply chain attacks include the SolarWinds, log4j, and xz utils incidents. Supply chain attacks are considered to have three major attack vectors: through vulnerabilities and malware accidentally or intentionally injected into open-source and third-party dependencies/components/containers ; by infiltrating the build infrastructure during the build and deployment processes; and through targeted techniques aimed at the humans involved in software development, such as through social engineering. Plummeting trust in the software supply chain could decelerate digital innovation if the software industry reduces its use of open-source and third-party artifacts to reduce risks. This paper contains perspectives and knowledge obtained from intentional outreach with practitioners to understand their practical challenges and from extensive research efforts. We then provide an overview of current research efforts to secure the software supply chain. Finally, we propose a future research agenda to close software supply chain attack vectors and support the software industry.

Disagree and commit: degrees of argumentation-based agreements

In cooperative human decision-making, agreements are often not total; a partial degree of agreement is sufficient to commit to a decision and move on, as long as one is somewhat confident that the involved parties are likely to stand by their commitment in the future, given no drastic unexpected changes. In this paper, we introduce the notion of agreement scenarios that allow artificial autonomous agents to reach such agreements, using formal models of argumentation, in particular abstract argumentation and value-based argumentation. We introduce the notions of degrees of satisfaction and (minimum, mean, and median) agreement, as well as a measure of the impact a value in a value-based argumentation framework has on these notions. We then analyze how degrees of agreement are affected when agreement scenarios are expanded with new information, to shed light on the reliability of partial agreements in dynamic scenarios. An implementation of the introduced concepts is provided as part of an argumentation-based reasoning software library.

Pilot functional magnetic resonance imaging study of clemastine‐aided functional recovery on a mouse model of demyelination

ABSTRACTBackgroundMultiple sclerosis (MS) is a chronic demyelinating disease characterized by autoimmune attacks on myelin sheaths. Its deleterious effects may be reversed by remyelination, a process that restores the integrity of myelin sheaths and, consequently, neuronal function. However, the functional implications of demyelination and remyelination in MS, as well as the potential impact of therapeutic interventions, remain incompletely understood. We used noninvasive longitudinal resting‐state functional magnetic resonance imaging in a cuprizone murine model of demyelination to investigate these unsolved questions.MethodsThree groups of (n = 6) animals were studied. A control group was fed with standard food for 5 weeks while two treatment groups (cuprizone and clemastine) suffered progressive demyelination by feeding them with 2% cuprizone. At Week 5 (W5), all animals returned to the standard diet and studied for another 5‐week period to compare controls vs spontaneous (cuprizone group) vs clemastine‐aided (clemastine group) remyelination group. Group clemastine was treated with this antihistaminic (oral gavage) during the remyelination period (Weeks 5–10). Anatomical magnetic resonance imaging (T2w‐MRI) and resting state functional MRI (rs‐FMRI) studies were conducted on weeks W0, W2, W5 (maximal demyelination) W7 and W10 (remyelination). MRI images were processed with the FMRIB Software Library, involving seed‐free functional imaging and seed‐based correlation. This study uses the t‐test and the D'Agostino–Pearson normality test to make an assessment.ResultsThe principal findings of our research include: (1) cuprizone‐treated animals suffer an initial phase of elevated connectivity at Week 2 with respect to controls, transitioning to reduced connectivity at Week 5; (2) different temporal trajectories across brain regions, reflecting varying susceptibility to demyelination; (3) while spontaneous remyelination normalizes connectivity in most networks at Week 10 (5 weeks after ceasing cuprizone intoxication), the thalamocortical axis exhibits lasting disruption even 6 months after normalization of diet; and (4) on the contrary, clemastine‐aided remyelination re‐establishes normal thalamocortical connectivity at 6 months after demyelination.ConclusionThis approach provides insights into the dynamic processes of demyelination and remyelination, informing the development of more effective interventions for MS.

An analysis of performance bottlenecks in MRI preprocessing.

Magnetic resonance imaging (MRI) preprocessing is a critical step for neuroimaging analysis. However, the computational cost of MRI preprocessing pipelines is a major bottleneck for large cohort studies and some clinical applications. While high-performance computing and, more recently, deep learning have been adopted to accelerate the computations, these techniques require costly hardware and are not accessible to all researchers. Therefore, it is important to understand the performance bottlenecks of MRI preprocessing pipelines to improve their performance. Using the Intel VTune profiler, we characterized the bottlenecks of several commonly used MRI preprocessing pipelines from the Advanced Normalization Tools (ANTs), FMRIB Software Library, and FreeSurfer toolboxes. We found few functions contributed to most of the CPU time and that linear interpolation was the largest contributor. Data access was also a substantial bottleneck. We identified a bug in the Insight Segmentation and Registration Toolkit library that impacts the performance of the ANTs pipeline in single precision and a potential issue with the OpenMP scaling in FreeSurfer recon-all. Our results provide a reference for future efforts to optimize MRI preprocessing pipelines.