Papers

Electrocardiogram (ECG) signals are often degraded by various noise sources such as baseline wander, motion artifacts, and electromyographic interference, posing a major challenge in clinical settings. This paper presents a lightweight deep learning-based denoising framework, forming a compact autoencoder architecture. The model was trained under severe noise conditions (-5 dB signal-to-noise ratio (SNR)) using a rigorously partitioned dataset to ensure no data leakage and robust generalization. Extensive evaluations were conducted across seven noise configurations and three SNR levels (-5 dB, 0 dB, and +5 dB), showing consistent denoising performance with minimal morphological distortion, critical for maintaining diagnostic integrity. In particular, tests on clinically vital rhythms such as ventricular tachycardia (VT) and ventricular fibrillation (VF) confirm that the proposed model effectively suppresses noise without altering arrhythmic features essential for diagnosis. Visual and quantitative assessments, including SNR improvement, RMSE, and correlation metrics, validate the model's efficacy in preserving waveform fidelity. To demonstrate real-world applicability, the model was deployed on a Raspberry Pi 4 using TensorFlow Lite with float16 precision. Inference latency was measured at just 1.41 seconds per 14-second ECG segment, indicating feasibility for near-real-time use in edge devices. Overall, this study introduces a lightweight, hardware-validated, and morphologically reliable ECG denoising solution suitable for integration into portable or wearable healthcare systems.

Next-generation wireless networks will rely on mmWave/sub-THz spectrum and extremely large antenna arrays (ELAAs). This will push their operation into the near field where far-field beam management degrades and beam training becomes more costly and must be done more frequently. Because ELAA training and data transmission consume energy and training trades off with service time, we pose a cross-layer control problem that couples PHY-layer beam management with MAC-layer service under delay-sensitive traffic. The controller decides when to retrain and how aggressively to train (pilot count and sparsity) while allocating transmit power, explicitly balancing pilot overhead, data-phase rate, and energy to reduce the queueing delay of MAC-layer frames/packets to be transmitted. We model the problem as a partially observable Markov decision process and solve it with deep reinforcement learning. In simulations with a realistic near-field channel and varying mobility and traffic load, the learned policy outperforms strong 5G-NR–style baselines at a comparable energy: it achieves 85.5% higher throughput than DFT sweeping and reduces the overflow rate by 78%. These results indicate a practical path to overhead-aware, traffic-adaptive near-field beam management with implications for emerging low-latency, high-rate next-generation applications such as digital twin, spatial computing, and immersive communication.

This letter studies the problem of online multi-step-ahead prediction for unknown linear stochastic systems. Using conditional distribution theory, we derive an optimal parameterization of the prediction policy as a linear function of future inputs, past inputs, and past outputs. Based on this characterization, we propose an online least-squares algorithm to learn the policy and analyze its regret relative to the optimal model-based predictor. We show that the online algorithm achieves logarithmic regret with respect to the optimal Kalman filter in the multi-step setting. Furthermore, with new proof techniques, we establish an almost-sure regret bound that does not rely on fixed failure probabilities for sufficiently large horizons $N$. Finally, our analysis also reveals that, while the regret remains logarithmic in $N$, its constant factor grows polynomially with the prediction horizon $H$, with the polynomial order set by the largest Jordan block of eigenvalue 1 in the system matrix.

Singular value decomposition (SVD) is widely used for dimensionality reduction and noise suppression, and it plays a pivotal role in numerous scientific and engineering applications. As the dimensions of the matrix grow rapidly, the computational cost increases significantly, posing a serious challenge to the efficiency of data analysis and signal processing systems, especially in time-sensitive scenarios involving large-scale datasets. Although various dedicated hardware architectures have been proposed to accelerate the computation of intensive SVD, many of these designs suffer from limited scalability and high consumption of on-chip memory resources. Moreover, they typically overlook the computational and data transfer challenges associated with SVD, making them unsuitable for real-time processing of large-scale data stream matrices in embedded systems. In this paper, we propose a Data Stream-Based SVD processing algorithm (DSB Jacobi), which significantly reduces on-chip BRAM usage while improving computational speed, offering a practical solution for real-time SVD computation of large-scale data streams. Compared to previous works, our experimental results indicate that the proposed method reduces on-chip RAM consumption by 41.5 percent and improves computational efficiency by a factor of 23.

X-ray scattering measurements of in situ human brain tissue encode structural signatures of pathological cross-$\beta$ inclusions, yet systematic exploitation of these data for automated detection remains challenging due to substrate contamination, strong inter-feature correlations, and limited sample sizes. This work develops a three-stage classification framework for identifying cross-$\beta$ structural inclusions-a hallmark of Alzheimer's disease-in X-ray scattering profiles of post-mortem human brain. Stage 1 employs a Bayes-optimal classifier to separate mica substrate from tissue regions on the basis of their distinct scattering signatures. Stage 2 introduces a multicollinearityaware, class-conditional correlation pruning scheme with formal guarantees on the induced Bayes risk and approximation error, thereby reducing redundancy while retaining class-discriminative information. Stage 3 trains a compact neural network on the pruned feature set to detect the presence or absence of cross-$\beta$ fibrillar ordering. The top-performing model, optimized with a composite loss combining Focal and Dice objectives, attains a test F1-score of 84.30% using 11 of 211 candidate features and 174 trainable parameters. The overall framework yields an interpretable, theory-grounded strategy for data-limited classification problems involving correlated, high-dimensional experimental measurements, exemplified here by X-ray scattering profiles of neurodegenerative tissue.

Achieving long-term safety in uncertain/extreme environments while accounting for human preferences remains a fundamental challenge for autonomous systems. Existing methods often trade off long-term guarantees for fast real-time control and cannot adapt to variability in human preferences or risk tolerance. To address these limitations, we propose a language-guided adaptive probabilistic safety certificate (PSC) framework that guarantees long-term safety for stochastic systems under environmental uncertainty while accommodating diverse human preferences. The proposed framework integrates natural-language inputs from users and Bayesian estimators of the environment into adaptive safety certificates that explicitly account for user preferences, system dynamics, and quantified uncertainties. Our key technical innovation leverages probabilistic invariance–a generalization of forward invariance to a probability space–to obtain myopic safety conditions with long-term safety guarantees. We validate the framework through numerical simulations of autonomous lane-keeping with human-in-the-loop guidance under uncertain and extreme road conditions, demonstrating enhanced safety-performance trade-offs, adaptability to changing environments, and personalization to different user preferences. Code is available at https://github.com/hoshino06/adaptive_lane_keeping.

Quantitative magnetization transfer (qMT) imaging provides myelin-sensitive biomarkers, such as the pool size ratio (PSR), which is valuable for multiple sclerosis (MS) assessment. However, qMT requires specialized 20-30 minute scans. We propose DEMIST to synthesize PSR maps from standard T1w and FLAIR images using a 3D latent diffusion model with three complementary conditioning mechanisms. Our approach has two stages: first, we train separate autoencoders for PSR and anatomical images to learn aligned latent representations. Second, we train a conditional diffusion model in this latent space on top of a frozen diffusion foundation backbone. Conditioning is decoupled into: (i) semantic tokens via cross-attention, (ii) spatial per-scale residual hints via a 3D ControlNet branch, and (iii) adaptive LoRA-modulated attention. We include edge-aware loss terms to preserve lesion boundaries and alignment losses to maintain quantitative consistency, while keeping the number of trainable parameters low and retaining the inductive bias of the pretrained model. We evaluate on 163 scans from 99 subjects using 5-fold cross-validation. Our method outperforms VAE, GAN and diffusion baselines on multiple metrics, producing sharper boundaries and better quantitative agreement with ground truth. Our code is publicly available at https://github.com/MedICL-VU/MS-Synthesis-3DcLDM.

Modern social networks rely on recommender systems that inadvertently amplify misinformation by prioritizing engagement over content veracity. We present a control framework that mitigates misinformation spread while maintaining user engagement by penalizing content characteristics commonly exploited by false information, specifically, extreme negative sentiment and novelty. We extend the closed-loop Friedkin-Johnsen model to incorporate the mitigation of misinformation together with the maximization of user engagement. Both model-free and model-based control strategies demonstrate up to 76% reduction in misinformation propagation across diverse network configurations, validated through simulations using the LIAR2 dataset with sentiment features extracted via large language models. Analysis of engagement-misinformation trade-offs reveals that in networks with radical users, median engagement improves even as misinformation decreases, suggesting content moderation enhances discourse quality for non-extremist users. The framework provides practical guidance for platform operators in balancing misinformation suppression with engagement objectives.

The interference of fluorescence signals and noise remains a significant challenge in Raman spectrum analysis, often obscuring subtle spectral features that are critical for accurate analysis. Inspired by variational methods similar to those used in image denoising, our approach minimizes a functional involving fractional derivatives to balance noise suppression with the preservation of essential chemical features of the signal, such as peak position, intensity, and area. The original problem is reformulated in the frequency domain through the Fourier transform, making the implementation simple and fast. In this work, we discuss the theoretical framework, practical implementation, and the advantages and limitations of this method in the context of {simulated} Raman data, as well as in image processing. The main contribution of this article is the combination of a variational approach in the frequency domain, the use of fractional derivatives, and the optimization of the {regularization parameter and} derivative order through the concept of Shannon entropy. This work explores how the fractional order, combined with the regularization parameter, affects noise removal and preserves the essential features of the spectrum {and image}. Finally, the study shows that the combination of the proposed strategies produces an efficient, robust, and easily implementable filter.

In this paper, we introduce OBHS (Optimized Block Huffman Scheme), a novel lossless audio compression algorithm tailored for real-time streaming applications. OBHS leverages block-wise Huffman coding with canonical code representation and intelligent fallback mechanisms to achieve high compression ratios while maintaining low computational complexity. Our algorithm partitions audio data into fixed-size blocks, constructs optimal Huffman trees for each block, and employs canonical codes for efficient storage and transmission. Experimental results demonstrate that OBHS attains compression ratios of up to 93.6% for silence-rich audio and maintains competitive performance across various audio types, including pink noise, tones, and real-world recordings. With a linear time complexity of O(n) for n audio samples, OBHS effectively balances compression efficiency and computational demands, making it highly suitable for resource-constrained real-time audio streaming scenarios.

Fiber Specklegram Sensors (FSS) are highly effective for environmental monitoring, particularly for detecting temperature variations. However, the nonlinear nature of specklegram data presents significant challenges for accurate temperature prediction. This study investigates the use of transformer-based architectures, including Vision Transformers (ViTs), Swin Transformers, and emerging models such as Learnable Importance Non-Symmetric Attention Vision Transformers (LINA-ViT) and Multi-Adaptive Proximity Vision Graph Attention Transformers (MAP-ViGAT), to predict temperature from specklegram data over a range of 0 to 120 Celsius. The results show that ViTs achieved a Mean Absolute Error (MAE) of 1.15, outperforming traditional models such as CNNs. GAT-ViT and MAP-ViGAT variants also demonstrated competitive accuracy, highlighting the importance of adaptive attention mechanisms and graph-based structures in capturing complex modal interactions and phase shifts in specklegram data. Additionally, this study incorporates Explainable AI (XAI) techniques, including attention maps and saliency maps, to provide insights into the decision-making processes of the transformer models, improving interpretability and transparency. These findings establish transformer architectures as strong benchmarks for optical fiber-based temperature sensing and offer promising directions for industrial monitoring and structural health assessment applications.

Renewable charging stations (RCSs) that co-locate electric-vehicle (EV) charging with distributed generation (DG) can raise renewable utilization and improve distribution-network (DN) efficiency, yet their variability and the siting-dependent charging demand can overload feeders if placed poorly. This paper proposes a tri-level, two-stage stochastic-robust optimization (SRO) co-planning framework that jointly determines RCS siting and DN expansion while accounting for transportation flows and population density. The model distinguishes two uncertainty classes: (i) decision-dependent uncertainty (DDU), under which EV charging loads vary with RCS siting; and (ii) decision-independent uncertainty (DIU), under which load fluctuations and renewable-generation variability do not depend on the RCS locations or the DN topology. At the upper level, the framework selects RCS sites and DN expansions. At the middle level, EV routing and charging are dispatched given the RCS siting to produce charging loads DDU. At the lower level, DN operation minimizes annualized loss costs under the worst-case DIU, reformulated via Karush-Kuhn-Tucker (KKT) conditions. To solve the resulting problem efficiently, we develop an adaptive inexact column-and-constraint generation (A-iC&CG) algorithm and prove finite-iteration convergence. Case studies on a 47-node DN coupled with a 68-hub transportation network in Shenzhen, China, show that A-iC&CG outperforms benchmark algorithms and that PV-EV hybrid stations are cost-optimal, with RCS siting concentrated near substations and high-flow hubs.

In the day-ahead energy market, the offering strategy of distributed energy resource (DER) aggregators must be submitted before the uncertainty realization in the form of price-quantity pairs. This work addresses the day-ahead offering problem through a two-stage adaptive robust stochastic optimization model, wherein the first-stage price-quantity pairs and second-stage operational commitment decisions are made before and after DER uncertainty is realized, respectively. Uncertainty in day-ahead price is addressed using a stochastic programming-based approach, while uncertainty of DER generation is handled through robust optimization. To address the max-min structure of the second-stage problem, a neural network-accelerated column-and-constraint generation method is developed. A dedicated neural network is trained to approximate the value function, while optimality is maintained by the design of the network architecture. Numerical studies indicate that the proposed method yields high-quality solutions and is up to 100 times faster than Gurobi and 33 times faster than classical column-and-constraint generation on the same 1028-node synthetic distribution network.

In the field of medical imaging, AI-assisted techniques such as object detection, segmentation, and classification are widely employed to alleviate the workload of physicians and doctors. However, single-task models are predominantly used, overlooking the shared information across tasks. This oversight leads to inefficiencies in real-life applications. In this work, we propose MTMed3D, a novel end-to-end Multi-task Transformer-based model to address the limitations of single-task models by jointly performing 3D detection, segmentation, and classification in medical imaging. Our model uses a Transformer as the shared encoder to generate multi-scale features, followed by CNN-based task-specific decoders. The proposed framework was evaluated on the BraTS 2018 and 2019 datasets, achieving promising results across all three tasks, especially in detection, where our method achieves better results than prior works. Additionally, we compare our multi-task model with equivalent single-task variants trained separately. Our multi-task model significantly reduces computational costs and achieves faster inference speed while maintaining comparable performance to the single-task models, highlighting its efficiency advantage. To the best of our knowledge, this is the first work to leverage Transformers for multi-task learning that simultaneously covers detection, segmentation, and classification tasks in 3D medical imaging, presenting its potential to enhance diagnostic processes. The code is available at https://github.com/fanlimua/MTMed3D.git.

Hybrid dynamical systems result from the interaction of continuous-variable dynamics with discrete events and encompass various systems such as legged robots, vehicles and aircrafts. Challenges arise when the system's modes are characterized by unobservable (latent) parameters and the events that cause system dynamics to switch between different modes are also unobservable. Model-based control approaches typically do not account for such uncertainty in the hybrid dynamics, while standard model-free RL methods fail to account for abrupt mode switches, leading to poor generalization. To overcome this, we propose SAC-MoE which models the actor of the Soft Actor-Critic (SAC) framework as a Mixture-of-Experts (MoE) with a learned router that adaptively selects among learned experts. To further improve robustness, we develop a curriculum-based training algorithm to prioritize data collection in challenging settings, allowing better generalization to unseen modes and switching locations. Simulation studies in hybrid autonomous racing and legged locomotion tasks show that SAC-MoE outperforms baselines (up to 6x) in zero-shot generalization to unseen environments. Our curriculum strategy consistently improves performance across all evaluated policies. Qualitative analysis shows that the interpretable MoE router activates different experts for distinct latent modes.

This paper tackles the sensing-communication trade-off in integrated sensing and communication (ISAC)-empowered subnetworks for mono-static target localization. We propose a low-complexity iterative node selection algorithm that exploits the spatial diversity of subnetwork deployments and dynamically refines the set of sensing subnetworks to maximize localization accuracy under tight resource constraints. Simulation results show that our method achieves sub-7 cm accuracy in additive white Gaussian noise (AWGN) channels within only three iterations, yielding over 97% improvement compared to the best-performing benchmark under the same sensing budget. We further demonstrate that increasing spatial diversity through additional antennas and subnetworks enhances sensing robustness, especially in fading channels. Finally, we quantify the sensing-communication trade-off, showing that reducing sensing iterations and the number of sensing subnetworks improves throughput at the cost of reduced localization precision.

We introduce VoiceCraft-X, an autoregressive neural codec language model which unifies multilingual speech editing and zero-shot Text-to-Speech (TTS) synthesis across 11 languages: English, Mandarin, Korean, Japanese, Spanish, French, German, Dutch, Italian, Portuguese, and Polish. VoiceCraft-X utilizes the Qwen3 large language model for phoneme-free cross-lingual text processing and a novel token reordering mechanism with time-aligned text and speech tokens to handle both tasks as a single sequence generation problem. The model generates high-quality, natural-sounding speech, seamlessly creating new audio or editing existing recordings within one framework. VoiceCraft-X shows robust performance in diverse linguistic settings, even with limited per-language data, underscoring the power of unified autoregressive approaches for advancing complex, real-world multilingual speech applications. Audio samples are available at https://zhishengzheng.com/voicecraft-x/.

We consider a variant of the target defense problem where a single defender is tasked to capture a sequence of incoming intruders. Both the defender and the intruders have non-holonomic dynamics. The intruders' objective is to breach the target perimeter without being captured by the defender, while the defender's goal is to capture as many intruders as possible. After one intruder breaches or is captured, the next appears randomly on a fixed circle surrounding the target. Therefore, the defender's final position in one game becomes its starting position for the next. We divide an intruder-defender engagement into two phases, partial information and full information, depending on the information available to the players. We address the capturability of an intruder by the defender using the notions of Dubins path and guarding arc. We quantify the percentage of capture for both finite and infinite sequences of incoming intruders. Finally, the theoretical results are verified through numerical examples using Monte-Carlo-type random trials of experiments.

This paper investigates the sensing potential of affine frequency division multiplexing (AFDM) in high-mobility integrated sensing and communication (ISAC) from the perspective of radar waveforms. We introduce an innovative parameter selection criterion that establishes a precise mathematical equivalence between AFDM subcarriers and Nyquist-sampled frequency-modulated continuous-wave (FMCW). This connection not only provides a clear physical insight into AFDM's sensing mechanism but also enables a direct mapping from the DAFT index to delay-Doppler (DD) parameters of wireless channels. Building on this, we develop a novel input-output model in a DD-parameterized DAFT (DD-DAFT) domain for AFDM, which explicitly reveals the inherent DD coupling effect arising from the chirp-channel interaction. Subsequently, we design two matched-filtering sensing algorithms. The first is performed in the time-frequency domain with low complexity, while the second is operated in the DD-DAFT domain to precisely resolve the DD coupling. Simulations show that our algorithms achieve effective pilot-free sensing and demonstrate a fundamental trade-off between sensing performance, communication overhead, and computational complexity. The proposed AFDM outperforms classical AFDM and other variants in most scenarios.

Oscillatory dynamics have recently proven highly effective in machine learning (ML), particularly through State-Space-Models (SSM) that leverage structured linear recurrences for long-range temporal processing. Resonate-and-Fire neurons capture such oscillatory behavior in a spiking framework, offering strong expressivity with sparse event-based communication. While early analog RAF circuits employed nonlinear coupling and suffered from process sensitivity, modern ML practice favors linear recurrence. In this work, we introduce a resonate-and-fire (RAF) neuron, built in 22nm Fully-Depleted Silicon-on-Insulator technology, that aligns with SSM principles while retaining the efficiency of spike-based communication. We analyze its dynamics, linearity, and resilience to Process, Voltage, and Temperature variations, and evaluate its power, performance, and area trade-offs. We map the characteristics of our circuit into a system-level simulation where our RAF neuron is utilized in a keyword-spotting task, showing that its non-idealities do not hinder performance. Our results establish RAF neurons as robust, energy-efficient computational primitives for neuromorphic hardware.

Lexical tone is central to many languages but remains underexplored in self-supervised learning (SSL) speech models, especially beyond Mandarin. We study four languages with complex and diverse tone systems (Burmese, Thai, Lao, and Vietnamese) to ask how far such models "listen" for tone and how transfer operates in low-resource conditions. As a baseline reference, we estimate the temporal span of tone cues: approximately 100ms (Burmese/Thai) and 180ms (Lao/Vietnamese). Probes and gradient analysis on fine-tuned SSL models reveal that tone transfer varies by downstream task: automatic speech recognition fine-tuning aligns spans with language-specific tone cues, while prosody- and voice-related tasks bias toward overly long spans. These findings indicate that tone transfer is shaped by downstream task, highlighting task effects on temporal focus in tone modeling.

The proliferation of SQL for data processing has often occurred without the rigor of traditional software development, leading to siloed efforts, logic replication, and increased risk. This ad-hoc approach hampers data governance and makes validation nearly impossible. Organizations are adopting DataOps, a methodology combining Agile, Lean, and DevOps principles to address these challenges to treat analytics pipelines as production systems. However, a standardized framework for implementing DataOps is lacking. This perspective proposes a qualitative design for a DataOps-aligned validation framework. It introduces a DataOps Controls Scorecard, derived from a multivocal literature review, which distills key concepts into twelve testable controls. These controls are then mapped to a modular, extensible CI/CD pipeline framework designed to govern a single source of truth (SOT) SQL repository. The framework consists of five stages: Lint, Optimize, Parse, Validate, and Observe, each containing specific, automated checks. A Requirements Traceability Matrix (RTM) demonstrates how each high-level control is enforced by concrete pipeline checks, ensuring qualitative completeness. This approach provides a structured mechanism for enhancing data quality, governance, and collaboration, allowing teams to scale analytics development with transparency and control.

Cytology is a valuable tool for early detection of oral squamous cell carcinoma (OSCC). However, manual examination of cytology whole slide images (WSIs) is slow, subjective, and depends heavily on expert pathologists. To address this, we introduce the first weakly supervised deep learning framework for patient-level diagnosis of oral cytology whole slide images, leveraging the newly released Oral Cytology Dataset [1], which provides annotated cytology WSIs from ten medical centres across India. Each patient case is represented as a bag of cytology patches and assigned a diagnosis label (Healthy, Benign, Oral Potentially Malignant Disorders (OPMD), OSCC) by an in-house expert pathologist. These patient-level weak labels form a new extension to the dataset. We evaluate a baseline multiple-instance learning (MIL) model and a proposed Region-Affinity Attention MIL (RAA-MIL) that models spatial relationships between regions within each slide. The RAA-MIL achieves an average accuracy of 72.7%, weighted F1-score of 0.69 on an unseen test set, outperforming the baseline. This study establishes the first patient-level weakly supervised benchmark for oral cytology and moves toward reliable AI-assisted digital pathology.

Early detection of oral cancer and potentially malignant diseases is a major challenge in low-resource settings due to the scarcity of annotated data. We provide a unified approach for oral lesion classification that incorporates deep learning, spectral analysis, and demographic data. A pathologist verified subset of oral cavity images was curated from a publicly available dataset. Oral cavity pictures were processed using a fine tuned ConvNeXtv2 network for deep embeddings before being translated into the hyperspectral domain using a reconstruction algorithm. Haemoglobin sensitive, textural, and spectral descriptors were obtained from the reconstructed hyperspectral cubes and combined with demographic data. Multiple machine learning models were evaluated using patient specific validation. Finally, an incremental heuristic meta learner (IHML) was developed that merged calibrated base classifiers via probabilistic feature stacking and uncertainty-aware abstraction of multimodal representations with patient level smoothing. By decoupling evidence extraction from decision fusion, IHML stabilizes predictions in heterogeneous, small sample medical datasets. On an unseen test set, our proposed model achieved a macro F1 of 66.23% and an overall accuracy of 64.56%. The findings demonstrate that RGB to hyperspectral reconstruction and ensemble meta learning improve diagnostic robustness in real world oral lesion screening.

This paper proposes a novel Bayesian framework for solving Poisson inverse problems by devising a Monte Carlo sampling algorithm which accounts for the underlying non-Euclidean geometry. To address the challenges posed by the Poisson likelihood – such as non-Lipschitz gradients and positivity constraints – we derive a Bayesian model which leverages exact and asymptotically exact data augmentations. In particular, the augmented model incorporates two sets of splitting variables both derived through a Bregman divergence based on the Burg entropy. Interestingly the resulting augmented posterior distribution is characterized by conditional distributions which benefit from natural conjugacy properties and preserve the intrinsic geometry of the latent and splitting variables. This allows for efficient sampling via Gibbs steps, which can be performed explicitly for all conditionals, except the one incorporating the regularization potential. For this latter, we resort to a Hessian Riemannian Langevin Monte Carlo (HRLMC) algorithm which is well suited to handle priors with explicit or easily computable score functions. By operating on a mirror manifold, this Langevin step ensures that the sampling satisfies the positivity constraints and more accurately reflects the underlying problem structure. Performance results obtained on denoising, deblurring, and positron emission tomography (PET) experiments demonstrate that the method achieves competitive performance in terms of reconstruction quality compared to optimization- and sampling-based approaches.

We propose a prompt-conditioned framework built on MedSigLIP that injects textual priors via Feature-wise Linear Modulation (FiLM) and multi-scale pooling. Text prompts condition patch-token features on clinical intent, enabling data-efficient learning and rapid adaptation. The architecture combines global, local, and texture-aware pooling through separate regression heads fused by a lightweight MLP, trained with pairwise ranking loss. Evaluated on the LDCTIQA2023 (a public LDCT quality assessment challenge) with 1,000 training images, we achieve PLCC = 0.9575, SROCC = 0.9561, and KROCC = 0.8301, surpassing the top-ranked published challenge submissions and demonstrating the effectiveness of our prompt-guided approach.

Block-Matching and 3D Filtering (BM3D) exploits non-local self-similarity priors for denoising but relies on fixed parameters. Deep models such as U-Net are more flexible but often lack interpretability and fail to generalize across noise regimes. In this study, we propose Deep Unfolded BM3D (DU-BM3D), a hybrid framework that unrolls BM3D into a trainable architecture by replacing its fixed collaborative filtering with a learnable U-Net denoiser. This preserves BM3D's non-local structural prior while enabling end-to-end optimization. We evaluate DU-BM3D on low-dose CT (LDCT) denoising and show that it outperforms classic BM3D and standalone U-Net across simulated LDCT at different noise levels, yielding higher PSNR and SSIM, especially in high-noise conditions.

Modern deep learning systems are typically deployed as open-loop function approximators: they map inputs to outputs in a single pass, without regulating how much computation or explanatory effort is spent on a given case. In safety-critical settings, this is brittle: easy and ambiguous inputs receive identical processing, and uncertainty is only read off retrospectively from raw probabilities. We introduce the Surgical Cognitive Interpreter (SCI), a lightweight closed-loop metacognitive control layer that wraps an existing stochastic model and turns prediction into an iterative process. SCI monitors a scalar interpretive state SP(t), here instantiated as a normalized entropy-based confidence signal, and adaptively decides whether to stop, continue sampling, or abstain. The goal is not to improve accuracy per se, but to regulate interpretive error {\Delta}SP and expose a safety signal that tracks when the underlying model is likely to fail. We instantiate SCI around Monte Carlo dropout classifiers in three domains: vision (MNIST digits), medical time series (MIT-BIH arrhythmia), and industrial condition monitoring (rolling-element bearings). In all cases, the controller allocates more inference steps to misclassified inputs than to correct ones (up to about 3-4x on MNIST and bearings, and 1.4x on MIT-BIH). The resulting {\Delta}SP acts as a usable safety signal for detecting misclassifications (AUROC 0.63 on MNIST, 0.70 on MIT-BIH, 0.86 on bearings). Code and reproducibility: https://github.com/vishal-1344/sci

Particle filters (PFs) are often combined with swarm intelligence (SI) algorithms, such as Chicken Swarm Optimization (CSO), for particle rejuvenation. Separately, Kullback–Leibler divergence (KLD) sampling is a common strategy for adaptively sizing the particle set. However, the theoretical interaction between SI-based rejuvenation kernels and KLD-based adaptive sampling is not yet fully understood. This paper investigates this specific interaction. We analyze, under a simplified modeling framework, the effect of the CSO rejuvenation step on the particle set distribution. We propose that the fitness-driven updates inherent in CSO can be approximated as a form of mean-square contraction. This contraction tends to produce a particle distribution that is more concentrated than that of a baseline PF, or in mathematical terms, a distribution that is plausibly more “peaked” in a majorization sense. By applying Karamata's inequality to the concave function that governs the expected bin occupancy in KLD-sampling, our analysis suggests a connection: under the stated assumptions, the CSO-enhanced PF (CPF) is expected to require a lower expected particle count than the standard PF to satisfy the same statistical error bound. The goal of this study is not to provide a fully general proof, but rather to offer a tractable theoretical framework that helps to interpret the computational efficiency empirically observed when combining these techniques, and to provide a starting point for designing more efficient adaptive filters.

This paper studies the removal of salt-and-pepper noise from images using median filter (MF) and simple three-layer autoencoder (AE) within recursive threshold algorithm. The performance of denoising is assessed with two metrics: the standard Structural Similarity Index SSIMImg of restored and clean images and a newly applied metric SSIMMap - the SSIM of entropy maps of these images computed via 2D Sample Entropy in sliding windows. We shown that SSIMMap is more sensitive to blur and local intensity transitions and complements SSIMImg. Experiments on low- and high-resolution grayscales images demonstrate that recursive threshold MF robustly restores images even under strong noise (50-60 %), whereas simple AE is only capable of restoring images with low levels of noise (<30 %). We propose two scalable schemes: (i) 2MF, which uses two MFs with different window sizes and a final thresholding step, effective for highlighting sharp local details at low resolution; and (ii) MFs-AE, which aggregates features from multiple MFs via an AE and is beneficial for restoring the overall scene structure at higher resolution. Owing to its simplicity and computational efficiency, MF remains preferable for deployment on resource-constrained platforms (edge/IoT), whereas AE underperforms without prior denoising. The results also validate the practical value of SSIMMap for objective blur assessment and denoising parameter tuning.