Children and adults can make more efficient use of their limited working memory capacity by reorganizing information based on prior knowledge or perceptual characteristics. The data compression approach has proven useful to relate both ways of reorganizing information because it can account for how patterns can be optimally recoded. However, this approach has never been applied to the study of how basic patterns are encoded by children. The purpose of the present study was to examine whether a compressibility metric could predict working memory performance of children aged 3 to 5 years. We present one experiment in which the level of compressibility of a sequence of colors to remember was manipulated in an immediate serial recall task. Results from a large sample of 541 children suggest that young children reorganize information on the fly, but not entirely as predicted by a compressibility metric. We discuss how more general perceptual characteristics, such as symmetry and alternation, may complement formal complexity metrics in explaining children's recall patterns.

Three-dimensional (3D) imaging underpins applications ranging from autonomous navigation to defense and biomedicine, with single-photon avalanche diode (SPAD) light detection and ranging (LiDAR) enabling fast, long-range, and photon-efficient depth sensing. In practice, reconstruction quality is constrained by limited sensor resolution, particularly in the short- and medium-wave infrared, as well as sensor dark counts, background illumination, atmospheric effects, and motion. We introduce a unified framework for continuous-surface 3D scene representation that integrates multimodal sensing with score-based priors on latent variables. The proposed approach models scenes using a parametric continuous surface, enabling robust rendering at arbitrary spatial resolutions, even in the presence of multiple depth layers allowing imaging through camouflage. We demonstrate capabilities including data compression, targeted high-resolution rendering, and guided super-resolution of dynamic 3D videos. Validated across multiple sensing scenarios using diverse off-the-shelf priors, this framework enables compressed, high-fidelity 3D imaging in real-world environments.

Individuals with severe neurological injuries often rely on assistive technologies, but current methods have limitations in accurately decoding multi-degree-of-freedom (DoF) movements. Intracortical brain-machine interfaces (iBMIs) use neural signals to provide a more natural control method but currently struggle with higher-DoF movements: something the brain handles effortlessly. It has been theorized that the brain simplifies high-DoF movement through muscle synergies, which link multiple muscles to function as a single unit. These synergies have been studied using dimensionality reduction techniques like principal component analysis (PCA), non-negative matrix factorization (NMF), and demixed PCA (dPCA) and successfully used to reduce noise and improve offline decoder stability in non-invasive applications. However, their effectiveness in improving decoding and generalizability for implanted recordings across varied tasks is unclear. Here, we evaluated whether brain and muscle synergies can enhance iBMI performance in non-human primates performing a two-DoF finger task. Specifically, we tested if PCA, dPCA, and NMF could compress and denoise brain and muscle data and improve decoder generalization across tasks. Our results showed that while all methods effectively compressed data with minimal loss in decoding accuracy, none improved performance through denoising in our datasets. Additionally, none of the methods enhanced generalization across tasks. These findings suggest that while dimensionality reduction can aid data compression, extracting synergies alone did not provide an advantageous or cleaner control space for linear decoding in our study. Further research with larger sample sizes and more channels in muscle recordings is required to determine whether synergies can be leveraged as an optimal control framework or if alternative approaches are required to enhance decoder robustness in iBMI applications.

Quantitative analysis of nitrogen and phosphorus in dairy cow slurry based on near-infrared spectroscopy combined with deep learning and characteristic selection algorithm.

MotivationEfficient data compression is crucial for reducing storage and transmission costs associated to vast volumes of nanopore raw sequencing data. Surpassing the state-of-the-art compression performance has been challenging, and all recent progress in this direction either incur a computational performance over-cost or resort to lossy compression schemes, which are not always desirable.ResultsIn this article, we present PDZ, a lossless compression algorithm that outperforms VBZ, the current defacto standard, both in compression performance and computational efficiency. In our experimental evaluation, the compression ratio improvement ranges from 0.87% to 2.84% depending on the dataset, the compression speed is 1.09× to 2.25× faster depending on the hardware, and the decompression speed is 1.01× to 1.52× faster depending on the hardware. Compared to EX-ZD, a compression algorithm with similar compression performance, the speedup factor for both compression and decompression goes from approximately to , depending on the hardware.Availability and implementationPDZ is implemented in C++ as a new compression method within the POD5 format. The source code is available as a fork of the open-source NanoporeTech library at https://github.com/Rafael-Cast/Piecewise-Differential-Zstd-Coder-POD5-Demo.

The rapid development of multimodal epidermal sensing requires scalable, energy-efficient data processing architectures capable of processing large volumes of raw data. Conventional systems suffer from high energy consumption and transmission latency due to the physical separation of sensors and processors. Here, we present an ultrathin flexible edge computing circuit based on carbon nanotube thin-film transistors (CNT-TFTs) and machine learning (ML)-assisted design. By incorporating substrate engineering, ML-derived device modeling, and industry-compatible design methodologies, we establish a complete toolchain from device to system. The ML model achieves 91.2% prediction accuracy, enabling simulation-guided optimization of logic gates. A CNT-based standard cell library enables the construction of flexible circuits with 361 transistors and 160 logic gates. Monolithic integration with an 8-channel tilt sensor achieves 62.5% data compression while maintaining functionality after undergoing 360° deformation. This work establishes an ML-assisted CNT circuit design framework for fully integrated flexible edge computing, enabling scalable wearable applications.

This paper proposes a federated learning (FL) framework that incorporates adaptive gradient compression and dynamic aggregation to address communication efficiency and data privacy issues in the context of FL with limited sample size and non-IID data distributions in edge devices and resource-scarce environments. This proposed framework incorporates dynamic gradient compression techniques that compress gradients based on their magnitude and variance to ensure high communication efficiency with minimal loss in model accuracy. Meanwhile, the proposed framework incorporates dynamic aggregation techniques that assign different weights to clients based on their reliability to ensure effective model convergence in heterogeneous and scarce data distributions. Data privacy in the proposed framework is ensured through secure aggregation and Differential Privacy (DP) techniques. Experimental results on various datasets, including LEAF, FEMNIST, Reddit, and Shakespeare, show that the proposed framework ensures communication efficiency of over 70%, preserves model accuracy with minimal loss at 1–2%, and achieves 30% faster convergence speed compared to traditional FL techniques. These results show that the proposed framework is applicable in real-world scenarios in mobile edge computing and IoT applications, where communication efficiency and data privacy are significant factors for model convergence and deployment. The combination of gradient compression and dynamic aggregation in FL with strong privacy guarantees makes this framework a powerful tool for model convergence in heterogeneous scenarios.

Thermal constraint assessment method for mission verification of sun-synchronous orbit satellites

Scanpath prediction in panoramic videos is a challenging task due to the spherical geometry and multimodality of the input, and the inherent uncertainty and diversity of the output. To give a complete treatment of these characteristics, we first present a simple criterion for scanpath prediction based on principles from lossy data compression. This criterion suggests minimizing the expected code length of quantized scanpaths, corresponding to fitting a discrete conditional probability model via maximum likelihood. We condition the probability model on two modalities: a viewport sequence as the deformation-reduced visual input and a set of relative past scanpaths projected onto respective viewports as the aligned path input. Furthermore, we parameterize it by a product of discretized Gaussian mixture models to capture the uncertainty and diversity of scanpaths from different humans. In doing so, the training of the probability model does not rely on the specification of "ground-truth" scanpaths for imitation learning. We also introduce a proportional-integral-derivative (PID) controller-based sampler to generate realistic human-like scanpaths from the learned probability model. Experimental results demonstrate that our method consistently produces better quantitative scanpath results in terms of prediction accuracy (by comparing to the assumed "ground-truths") and perceptual realism (through machine discrimination) over a wide range of prediction horizons. We additionally verify the perceptual realism improvement via a formal psychophysical experiment and the generalization improvement on several unseen panoramic video datasets.

The rapid expansion of wearable sensing technologies and patient-reported outcomes (PROs) has revealed a persistent gap between high-frequency, context-rich patient-generated health data (PGHD) and standardized data models used for secondary analysis. Although the OMOP Common Data Model (OMOP-CDM) supports large-scale observational research, it does not natively capture key PGHD characteristics such as device provenance, calibration parameters, temporal resolution, and contextual metadata. This paper introduces an extension layer for OMOP-CDM and a direct extract-transform-load (ETL) pathway that enhances its capacity to handle heterogeneous, time-series, and PRO data while maintaining semantic consistency. The approach preserves device traceability, enables flexible data compression, and links PRO instruments with OMOP constructs without altering the core schema, remaining fully compatible with OHDSI analytical tools and aligned with FAIR principles. The model has strong potential for community discussion and future OMOP updates on PGHD.

Hyperspectral remote sensing images provide rich spatial and spectral information about the Earth’s surface, making them an essential tool for Earth observation. However, existing spaceborne hyperspectral payloads experience slow acquisition speeds and generate large data volumes, posing significant challenges for real-time applications. Moreover, the complex optical design and relatively high cost of traditional hyperspectral payloads hinder their broad-scale in-orbit deployment. In this work, we have proposed and completed the world’s first computational imaging-enabled compact spaceborne snapshot compressive hyperspectral payload, named BUPT-spectra01, which was successfully launched on November 11, 2024, at the Jiuquan Satellite Launch Center in China. We design a reflective coding structure, which enables BUPT-spectra01 to achieve high compactness (182 mm × 214 mm × 94 mm, 1.535 kg) and low cost. The payload operates in a sun-synchronous orbit at an altitude of 520 km, with a ground imaging swath width of 51 km by 64 km. Through a single exposure (1 ms), the payload enables 47-band hyperspectral imaging with a spectral resolution of 6.5 nm, achieving 47-times data compression simultaneously. To achieve high-accuracy hyperspectral information reconstruction, we design a novel spatial-spectral inference neural network (SSI-Net). Moreover, BUPT-spectra01 can image at a rate of 30 frames per second, which allows video-level hyperspectral observation. In-orbit experiments demonstrate that BUPT-spectra01 achieves accurate classification of ground cover based on hyperspectral features, showing promise in hyperspectral observation applications such as disaster management, environment monitoring, and resource exploration. This breakthrough significantly advances the application of computational imaging in aerospace observation, contributing to the progress of future satellite internet.

As a cornerstone of automated reasoning, equational reasoning finds equivalences between symbolic expressions and fuels advances across scientific disciplines. Yet, its potential remains limited by the exponential growth of equivalent expressions with increasing problem size. We introduce quantum normal form reduction, a quantum computational framework designed to address this challenge. We construct an efficiently implementable quantum Hamiltonian whose ground state encodes all equivalent expressions in a quantum superposition. By preparing and manipulating these states, we tackle fundamental problems in equational reasoning, including verifying and counting equivalent expressions and identifying structural properties of equivalence classes. We demonstrate a quantum-inspired version of the algorithm using tensor networks to solve instances involving up to 1028 equivalent expressions, far beyond the reach of classical graph exploration. This framework opens the path for quantum symbolic computation in areas from circuit design to data compression, computational group theory, linguistics, and macromolecular modeling, unlocking previously inaccessible problems.

Four-dimensional scanning transmission electron microscopy (4D-STEM) enables mapping of diffraction information with nanometer-scale spatial resolution, offering detailed insight into local structure, orientation, and strain. However, as data dimensionality and sampling density increase, particularly for in situ scanning diffraction experiments (5D-STEM), robust segmentation of structurally consistent behavior across sequential measurements becomes essential for efficient and physically meaningful analysis. Here, we introduce a clustering framework that identifies crystallographically distinct domains from 4D-STEM datasets. By using local diffraction-pattern similarity as a metric, the method extracts closed contours delineating spatially contiguous regions. This approach produces cluster-averaged diffraction patterns that improve signal quality while reducing data volume by orders of magnitude, enabling rapid and accurate orientation, phase, and strain mapping. We demonstrate the applicability of this approach to in situ liquid-cell 4D-STEM data of gold nanoparticle growth. Our method provides a scalable and generalizable route for spatially coherent segmentation, data compression, and quantitative structure-strain mapping across diverse 4D-STEM modalities. The full analysis code and example workflows are publicly available to support reproducibility and reuse.

Abstract The petabyte-scale data generated by High Energy Physics (HEP) experiments presents a significant storage challenge. We present the Bytewise Online Autoregressive (BOA) Constrictor, a new pseudo-streaming lossless neural compressor built upon the Mamba state space model. BOA achieves competitive compression ratios across diverse structured HEP datasets, matching or exceeding LZMA, ZSTD and ZLIB at maximum compression, among other tested algorithms. With a 2.21 MB model, BOA achieves an effective compression ratio (defined as the ratio of original to compressed file size, inclusive of model size) of 7.23× on ATLAS Open Data (HDF5) and 9.13× on simulated particle collision records (HepMC v3), outperforming the next-best traditional algorithm (6.79× and 5.33×, respectively on each dataset). BOA also demonstrates robust cross-file and cross-condition generalisation on CMS Open Data (NanoAOD format), where it obtains comparable or improved effective compression ratios (within 5%) with respect to the next-best traditional algorithm. Ablation studies show that transitioning to half-precision (FP16) weights reduces the model footprint without degrading predictive accuracy, and data-type analyses reveal BOA performs best on high-entropy float32 payloads. The model has also been tested in other kinds of scientific data, yielding 1.61× (vs. 1.14× for next-best algorithm) in computational fluid dynamics and up to 1.53× (vs. 1.27×) in cosmology (CAMELS) datasets. BOA is supported by a deterministic reference C++ implementation which ensures bit-exact reproducibility across different CUDA architectures. In this proof-of-principle implementation, BOA delivers a ∼3.5 to 45 MB/s compression and ∼1.5 to 25 MB/s decompression throughput that is not yet competitive with optimised algorithms such as ZSTD or LZMA, but still provides a first step towards data compression improvements for next-generation scientific data.

Data modeling in biomedical research often operates in the small-sample regime, where the number of observations is small relative to the data dimensionality; the detrimental effects of limited sample sizes are well documented in cancer studies. Synthetic data offers a potential solution to data shortfalls provided that the data generated is an adequate facsimile of the underlying distribution; the adequacy of such synthetic data remains an open-ended problem. In this work, we evaluate a synthetic generator proposed previously. The generator applies a series of transformations to the observed data to accommodate the small-sample size resulting in an uncoupled representation, where uncorrelated marginal distributions are modeled with optimized univariate kernel density estimation.In this report, (1) we develop a nonparametric method for assessing multivariate similarity based on the Cramér-Wold theorem and random projection testing, (2) investigate when the absence of bivariate correlation approximates independence in a non-normal setting, and (3) evaluate artifacts induced by data compression. The presentation is primarily methodological; low-dimensional data were used so each stage of the generation process could be analyzed explicitly.A formal testing framework was developed by comparing random projection level outcomes with a two-sample test, modeling these outcomes as Bernoulli trials, aggregating replicate outcomes within each projection direction, and pooling outcomes across many directions, yielding a scalable standardized normal test-statistic. The key innovation was decoupling the two-sample test significance level from that governing finalized normal inference. We showed the same projection framework also evaluates the full multivariate covariance structure. The generator produced high-fidelity multivariate synthetic data when the bivariate correlation approximates independence in the non-normal setting; in highly compressed data, residual modes were best modeled as normally distributed regardless of their intrinsic distributional form. Ongoing work includes applying these methods to higher-dimensional, diverse data.

Cardiovascular diseases (CVD) are among the leading causes of mortality worldwide due to genetic predisposition and lifestyle factors. Proper diagnosis of cardiovascular diseases is crucial to provide early-stage treatments. Conventional diagnostic methods such as stress tests, electrocardiograms, and echocardiography detect valuable insights into rhythm abnormalities, structural anomalies, or other cardiovascular conditions. However, their reliability heavily depends on human expertise, and they may not always detect early-stage signs of disease. In recent years, Machine Learning (ML) models have emerged as alternative diagnosis tools, capable of identifying CVD with higher accuracy. ML enables automated and precise detection based on data relationships, capturing hidden, complex patterns that are not apparent through traditional diagnostics. Most ML approaches employ supervised learning, which requires labeled data that are not always available in medical records. Under such circumstances, unsupervised learning has been explored as a suitable alternative. In this paper, a hybrid unsupervised approach combines the neural network structure of Self-Organizing Maps (SOM) with the dimensionality reduction technique of Principal Component Analysis (PCA) for unsupervised analysis for clustering CVD across different severity levels. Considering a data compression mechanism, the synergy among these methods leverages the ability to map unsupervised complex, high-dimensional data into lower-dimensional space. The proposed approach significantly improves the detection of hidden structures within large, high-dimensional medical cardiovascular datasets, providing insights into cardiovascular risk factors and improving the overall diagnostic process. Experimental evaluation on the UCI Cleveland Heart Disease dataset shows that the proposed PCA-SOM model achieves a Silhouette score of 0.94 (train) and 0.79 (test), and a Davies-Bouldin index of 0.08 (train) and 0.16 (test), outperforming baseline clustering methods such as K-means, hierarchical clustering, Gaussian Mixture and Spectral clustering highlighting its potential for supporting CVD detection.

Energy consumption is a critical concern for Internet of Things (IoT) platforms lacking abundant resources, particularly for swarm robotic systems that rely on numerous devices operating collaboratively over extended periods. This study presents a comprehensive design strategy for improving processing and communication to enhance system efficiency and reduce energy consumption. We incorporate energy harvesting (photovoltaic and RF), dynamic power management, and energy-efficient communication protocols (e.g., duty cycle, power control, data compression) into two complementary platforms built for swarm robotics: MCU-based nodes (TI MSP430 with LoRa transceiver), which serve as the experimental prototype for validating energy-aware communication, compression, and scheduling mechanisms; edge platforms (Jetson Nano and TX2), which are used for high-level power profiling and system-level evaluation, particularly for computation intensive workloads and comparative analysis. Our technique involves analyzing the device’s energy usage and harvesting processes, developing efficient communication protocols, and validating the system through simulations and hardware prototypes. Experimental results under outdoor and indoor conditions show that the device maintains an energy neutrality ratio well above unity, even with limited ambient energy. Key findings include significant reductions in energy per bit transmitted and reliable long-term operation. These insights pave the way for deploying swarms of autonomous IoT-based robots with minimal maintenance and maximal longevity.

A high-ratio lossy data compression framework for beam diagnostic systems using ResNet AutoEncoder networks

A test data compression method based on sliding-window encoding and matching length reuse

The prediction of Epidermal Growth Factor Receptor (EGFR) mutation status in advanced lung adenocarcinoma is crucial for targeted therapy. Since EGFR mutations manifest as both macroscopic imaging features on CT and microscopic morphological changes in tissue, integrating these multiscale signals is essential for a comprehensive diagnostic assessment. However, current related research faces two key limitations: on one hand, unimodal deep learning models suffer from limited representational power; on the other hand, existing multimodal methods fail to address the inherent data structural discrepancies between continuous CT and discrete WSI, often losing critical fine-grained details due to forced data compression or shared semantic bottlenecks. To address the above limitations and improve the reliability of EGFR mutation status prediction, this study aims to propose a novel multimodal fusion framework (MFCA) that can effectively capture cross-modal semantic interactions and align imaging features across different scales. A novel MFCA based on Cross-Attention (MFCA) is proposed, and its implementation steps are as follows: 1. First, a region-of-interest-guided approach is utilized to coarsely segment whole-slide histopathology images (WSI) into three constituent regions, namely cancerous, stromal, and other regions; 2. Then, a dual-branch encoder is employed to separately extract features from two types of imaging data-global features from Computed Tomography (CT) scans and region-specific features from the segmented WSI; 3. Critically, a bidirectional cross-attention module is introduced into the framework, which is designed to facilitate deep semantic interaction and alignment between the macroscopic context of CT imaging and the microscopic context of histopathology, thereby achieving highly efficient and discriminative feature fusion. On the external validation set, our MFCA framework achieved robust performance, with Area Under the Curve (AUC) values of 0.758(95% CI: 0.683-0.832) for cancerous regions, 0.805(95% CI: 0.716-0.900) for stromal regions, and 0.760(95% CI: 0.686-0.833) for other regions. The model's performance, particularly in the stromal component, was statistically superior to all baseline and competing models. The proposed MFCA framework predicts EGFR mutation status by innovatively integrating macroscopic CT imaging with region-specific microscopic WSI features. It serves as a valuable computational tool to support precision oncology for patients with advanced lung adenocarcinoma.