Empirical mode decomposition and interpretable machine learning for preterm birth classification from electrohysterography

Preterm birth (PTB) remains a major global health problem, and reliable non-invasive risk assessment remains difficult. Electrohysterography (EHG) records uterine electrical activity from the maternal abdomen and may support PTB assessment, but performance can be inflated when segments from the same recording are split across training and validation folds. We evaluated empirical mode decomposition (EMD) for term-versus-preterm classification using 26 pregnancy recordings (13 preterm, 13 term) from the public TPEHGT dataset. Annotated intervals and non-overlapping fixed 3-minute windows were compared, and the first four intrinsic mode functions (IMFs) were evaluated. Fourteen features from each of three EHG channels were assessed with nine classifiers using repeated five-fold recording-grouped cross-validation and recording-level aggregation. IMF1 gave the strongest mean performance. With fixed 3-minute IMF1 features, Random Forest achieved mean accuracy 0.8308, F1 0.7969, balanced accuracy 0.8308, MCC 0.6998, ROC-AUC 0.8157, and average precision 0.8877. IMF1 also outperformed matched filtered time-domain features across all reported mean metrics. Preterm recordings showed smaller, more regularly spaced peak-like events, lower temporal-energy measures, and higher entropy. These findings support further evaluation of IMF1-based EHG classification in larger independent cohorts.

Non-Destructive Quantification of Urea Adulteration in Bovine Milk Using Transmittance Multispectral Imaging

Adulteration of bovine milk using urea remains a major food quality and health concern, motivating the development of rapid and quantitative screening tools. Conventional approaches, including laboratory-based analytical methods and spectroscopic techniques, have been used for urea detection; however, many remain less suitable for rapid, low-cost routine screening due to requirements such as specialized instrumentation, sample preparation, chemical reagents, or laboratory operation. This study introduces a pragmatic, cost-effective, accurate, and laboratory-validated MSI-based method for quantitative urea estimation under controlled density conditions using a multispectral-imaging-based regression framework. An in-house-built multispectral imaging system operating in twelve discrete spectral bands (365--940~nm) was used to acquire multispectral images of milk samples prepared with controlled urea addition and water for density balancing. Fresh milk was obtained on the day of image acquisition, and the specific gravity of the milk was verified to be 1.032 at 20°C using a hydrometer. Multiple linear regression provided an initial mapping with a high validation $R^2$ of 0.9599, while a feed-forward neural network further improved predictive performance with a validation $R^2$ of 0.9773. These results demonstrate the feasibility of transmittance multispectral imaging for accurate, non-destructive urea quantification under controlled density-balanced conditions, supporting its potential as a rapid screening approach for milk-quality assessment.

A Synchronized Multi-IMU Wearable System for Tracking of Joint-Angles in Sports Motion Analysis With Reference-Based Validation and Dynamic Task Characterization

Reliable joint-angle measurement outside laboratory environments is important for posture assessment and technique analysis in sports and rehabilitation. Yet, wearable IMU systems, despite being the most practical solution, remain constrained by drift, multisensor timing consistency, validation, and cost. This paper presents a synchronized, cost-efficient, modular IMU wearable platform and an end-to-end pipeline for joint-angle estimation tailored to high-dynamic sports motion analysis. The system combines high-rate sensing with robust local logging, a Real Time Clock-disciplined microsecond timestamping scheme for inter-node synchronization, an indirect Kalman filter-based orientation estimator followed by relative-rotation joint-angle extraction, high-pass drift mitigation, and range normalization. Reference-based validation used a standardized seated knee flexion-extension protocol, comparing IMU-derived knee trajectories against a markerless vision reference computed from YOLOv11. The proposed method reproduced the expected 14-cycle motion and achieved low normalized error. Instrumentation performance was further evaluated using a long-duration (2h 12min) rigid-body elbow hold, yielding near-zero drift ($r_{\mathrm{drift}}=1.5\times10^{-6}$ deg/min) and a practical noise-limited resolution of $0.6442^\circ$. The system was also demonstrated on a high-dynamic clean & jerk task with two participants (professional and amateur) performing five consecutive lifts. Measured elbow and knee trajectories preserved stage-dependent signatures and rapid transients for technique interpretation, while enabling expertise-level discrimination through differences in transient timing and trajectory regularity. Overall, the results support the proposed system as a practical, temporally consistent, and cost-effective wearable approach for multi-joint technique analysis in high-dynamic settings.

A Controlled Visual-Backbone Benchmark for Multimodal Short-Term Solar Irradiance Forecasting

Sky-image irradiance studies often compare forecasting systems in which the image encoder, temporal model, fusion block, target definition, and training recipe all change together. We use a narrower protocol: the multimodal forecasting pipeline is fixed, and only the visual backbone is varied. The shared setup keeps preprocessing, clear-sky-index normalization, weather-history encoding, fusion, regression head, loss, optimizer schedule, seed, and chronological split policy unchanged. We compare ConvNeXt, Swin Transformer, VMamba, Spatial Mamba, and MambaVision backbones for 10min-ahead forecasting on Folsom and a strict matched NREL split. Forecast skill is measured against clear-sky-index smart persistence, and temporal-only rows are reported as weather-history diagnostics rather than as the main ranking criterion. On the Folsom strict split, all evaluated visual-backbone runs improve over smart persistence. In the evaluated single-seed strict runs, VMamba Small and Swin Base reach matched Folsom RMSE values of 65.39 W/m^2 and 65.50 W/m^2; the temporal-only diagnostic reaches 69.51 W/m^2. On the 313-sample NREL strict split, smart persistence remains strongest at 17.48 W/m^2, while the lowest visual RMSE is obtained by Swin Tiny at 23.76 W/m^2. These results provide a reproducible encoder comparison under one fixed multimodal operating point rather than establishing architecture-level dominance, statistically resolved ranking, or fully optimized forecasting performance. Code available here: https://github.com/Oshadha345/irradiance_benchmark

Structure over Depth: A Single-Block Spatio-Temporal Transformer for Multi-Entity Reasoning

Modeling multi-entity temporal data requires capturing dependencies across entities, time, and their interactions. Transformer-based approaches perform well but often rely on deep stacks of layers to learn these heterogeneous dependencies implicitly, increasing computational cost. We revisit this problem from a structural perspective and decompose multi-entity temporal dynamics into three interaction types: spatial interactions among entities, temporal interactions across time, and cross interactions coupling the two domains. We propose a structured spatio-temporal transformer block that explicitly models all three within a single stage. It uses parallel spatial and temporal self-attention, followed by bidirectional cross-attention, and combines the outputs through learnable gated fusion. By directly encoding these complementary views, the model reduces the need for deep stacking. We evaluate the approach on video-based group activity recognition, skeleton-based human interaction analysis, and wearable sensor-based activity recognition. Despite its simplicity, the single structured Transformer block matches or outperforms deeper architectures with only 1.76M parameters. The results suggest that depth in prior models partly compensates for implicit and entangled interaction modeling, whereas explicit factorization offers a more efficient and transparent alternative. More broadly, this work supports a structure-first design principle: expressive multi-entity temporal reasoning can emerge by exposing interaction structure rather than relying on depth.

MambaRefine-CD: MambaVision with Region-Boundary Temporal Refinement

Binary change detection in remote sensing requires both complete changed-region localization and accurate boundary delineation. We present MambaRefine-CD, a region-boundary temporal refinement framework built on a shared MambaVision encoder. The proposed D-RBI module constructs temporal evidence from paired features, absolute differences, and signed differences, then separates it into region and Sobel-conditioned boundary streams. Region features are enhanced with CRAM-lite and decoded by an adaptive receptive-field FPN, while the finest boundary stream guides a bounded residual refinement of the coarse prediction. Experiments on DSIFN-CD and WHU-CD show strong changed-class F1 and IoU under verified evaluation settings, and ablations support the contribution of signed temporal evidence and the full region-boundary refinement pipeline.

ScaLe-INR: Scale and Learn Implicit Neural Representations

Implicit Neural Representations (INRs) parameterized by multilayer perceptrons excel at modeling continuous signals. However, a key challenge persists as INRs fundamentally suffer from spectral bias and information cross-talk. When a single network attempts to capture multi-scale phenomena, high-frequency weight updates destructively interfere with the underlying low-frequency structural approximation. We introduce Scale and Learn INR (ScaLe-INR), a novel multi-branch architecture that resolves these limitations by explicitly matching the signal's frequency spectrum with the optimal operating region of the INR. Drawing upon the Fourier inverse scaling theorem we demonstrate that applying directional coordinate scaling expands a network's representational bandwidth along specific spatial axes. To mathematically enforce functional disentanglement and minimize task-specific information leakage between branches, we propose a Directional Edge Guidance Loss, a spatially-conditioned sparsity prior derived from ground-truth gradients. By constraining the high-frequency branches to act as strict, localized edge-filters, ScaLe-INR eliminates spectral cross-talk, accelerates convergence, and achieves high-fidelity signal reconstruction on complex multi-scale topologies. We evaluate ScaLe-INR across diverse reconstruction and inverse tasks, demonstrating substantial performance gains over existing state-of-the-art (SOTA) methods. The proposed architecture improves upon the nearest baselines by +5.16 dB in image reconstruction and +0.65 dB in image denoising. Furthermore, it achieve an impressive figure of 50.02 dB on audio reconstruction and 0.999 IOU(Intersection Over Union) on 3D reconstruction which beats the all SOTA models.

A Controlled Benchmark of Visual State-Space Backbones with Domain-Shift and Boundary Analysis for Remote-Sensing Segmentation

Visual state-space models (SSMs) are increasingly promoted as efficient alternatives to Vision Transformers, yet their practical advantages remain unclear under fair comparison because existing studies rarely isolate encoder effects from decoder and training choices. We present a strictly controlled benchmark of representative visual SSM families, including VMamba, MambaVision, and Spatial-Mamba, for remote-sensing semantic segmentation, in which only the encoder varies across experiments. Evaluated on LoveDA and ISPRS Potsdam under a unified 4-stage feature interface and a fixed lightweight decoder, the benchmark reveals three main findings, intra-family scaling yields only modest gains, cross-domain generalization is strongly asymmetric, and boundary delineation is the dominant failure mode under distribution shift. Although visual SSMs achieve favorable accuracy-efficiency trade-offs relative to the controlled CNN and Transformer baselines considered here, the results suggest that future improvements are more likely to come from robustness-oriented design and boundary-aware decoding than from encoder scaling alone. By isolating encoder behavior under a unified and reproducible protocol, this study establishes a practical reference benchmark for the design and evaluation of future Mamba-based segmentation backbones

Mitigating Long-Tail Bias via Prompt-Controlled Diffusion Augmentation

Semantic segmentation of high-resolution remote-sensing imagery is critical for urban mapping and land-cover monitoring, yet training data typically exhibits severe long-tailed pixel imbalance. In the dataset LoveDA, this challenge is compounded by an explicit Urban/Rural split with distinct appearance and inconsistent class-frequency statistics across domains. We present a prompt-controlled diffusion augmentation framework that synthesizes paired label--image samples with explicit control of both domain and semantic composition. Stage~A uses a domain-aware, masked ratio-conditioned discrete diffusion model to generate layouts that satisfy user-specified class-ratio targets while respecting learned co-occurrence structure. Stage~B translates layouts into photorealistic, domain-consistent images using Stable Diffusion with ControlNet guidance. Mixing the resulting ratio and domain-controlled synthetic pairs with real data yields consistent improvements across multiple segmentation backbones, with gains concentrated on minority classes and improved Urban and Rural generalization, demonstrating controllable augmentation as a practical mechanism to mitigate long-tail bias in remote-sensing segmentation. Source codes, pretrained models, and synthetic datasets are available at \href{https://github.com/Buddhi19/SyntheticGen.git}{Github}

Spectrotemporal Feature Extraction in EHG Signals and Tocograms for Enhanced Preterm Birth Prediction

Preterm birth (PTB), defined as delivery before 37 weeks of gestation, is a leading cause of neonatal mortality and long term health complications. Early detection is essential for enabling timely medical interventions. Electrohysterography (EHG) and tocography (TOCO) are promising non invasive tools for PTB prediction, but prior studies often suffer from class imbalance, improper oversampling, and reliance on features with limited physiological relevance. This work presents a machine learning pipeline incorporating robust preprocessing, physiologically grounded feature extraction, and rigorous evaluation. Features were extracted from EHG (and TOCO) signals using Mel frequency cepstral coefficients, statistical descriptors of wavelet coefficients, and peaks of the normalized power spectrum. Signal quality was enhanced via Karhunen Lo\`eve Transform (KLT) denoising through eigenvalue based subspace decomposition. Multiple classifiers, including Logistic Regression, Support Vector Machines, Random Forest, Gradient Boosting, Multilayer Perceptron, and CatBoost, were evaluated on the TPEHGT dataset. The CatBoost classifier with KLT denoising achieved the highest performance on fixed interval segments of the TPEHGT dataset, reaching 97.28% accuracy and an AUC of 0.9988. Ablation studies confirmed the critical role of both KLT denoising and physiologically informed features. Comparative analysis showed that including TOCO signals did not substantially improve prediction over EHG alone, highlighting the sufficiency of EHG for PTB detection. These results demonstrate that combining denoising with domain relevant features can yield highly accurate, robust, and clinically interpretable models, supporting the development of cost effective and accessible PTB prediction tools, particularly in low resource healthcare settings.