This retrospective single-center study developed and evaluated LiteFDNet, a lightweight deep neural network for intrapartum fetal distress detection from cardiotocography (CTG). All eligible deliveries from April 2014 to December 2024 were included under IRB approval (LGECH-2023-002; consent waived). Labels were refined using umbilical artery blood gas (distress: pH < 7.20; normal: pH 7.20–7.27), with exclusions for multiple gestations, major anomalies, missing CTG/outcomes, or non-analysable signals. Distress cases were matched 1:2 to controls, yielding 4,353 dual-channel CTG segments (1,451 distress; 2,902 normal). For each delivery, 30–60 min fetal heart rate and uterine contraction traces were artefact-screened, short gaps interpolated, z-normalised, lightly smoothed, and rendered as 224×224 images. LiteFDNet was inspired by ResNet-18 and trained with AdamW, cosine annealing, and mild affine augmentation; performance was primarily assessed using stratified 10-fold cross-validation, while an independent held-out test split was retained only for supplementary transparency analyses. The primary outcome was AUC, and secondary outcomes included accuracy, sensitivity, specificity, and F1 score. Among 4,353 women (maternal age 30.3 ± 4.4 years; gestational age 38.9 ± 1.6 weeks), LiteFDNet achieved a mean AUC of 0.940 ± 0.015 across the cross-validation folds and showed favorable performance relative to the compared baselines. Ablation analyses supported the contributions of the task-specific lightweight architectural modifications (t = 2.77; p < 0.01) and squeeze-and-excitation blocks (t = 2.15; p = 0.03). LiteFDNet may provide an accurate and computationally efficient proof-of-concept approach for CTG-based fetal distress screening, but further multi-center validation and comparison with stronger direct temporal baselines are warranted.
Washery middlings (WMC) are abundant by-products of coal preparation, yet their high ash content and associated safety risks hinder their large-scale application as blast-furnace injection fuels. In this study, we identify a beneficiation-induced quality-safety paradox and propose a coordinated regulation strategy that combines coal blending with carbonate additives. We find that while flotation upgrading successfully reduced the ash content from 23.70 wt% to 11.86 wt% - meeting the GB/T 18512-2022 specification - it selectively enriched reactive macerals and increased the volatile matter to 29.13 wt%. Consequently, long-tube explosion tests revealed a concerning shift, exacerbating the dust explosibility from weak to strong. To address this, particle-size control, anthracite (AC) blending, and the use of carbonate inhibitors were systematically evaluated. An optimal blend of 60 wt% WMC and 40 wt% anthracite effectively reverted the mixture to weak explosibility, primarily attributable to the physical heat-sink and barrier effects of the inert anthracite particles. Furthermore, thermogravimetric analyses demonstrated a non-linear, U-shaped variation in apparent activation energy (Ea) during cocombustion. The 60 wt% WMC blend exhibited the minimum Ea (99.74 kJ mol(-1)), indicating a heat-transferand radical-assisted synergy whereby volatile-rich WMC facilitates anthracite ignition. Trace CaCO3 at 2-4 wt % provided effective explosion mitigation while maintaining acceptable combustion performance, consistent with a stage-dependent balance between thermal inhibition and chemical promotion. Ultimately, an optimized injection formulation - 60 wt% WMC blended with 40 wt% anthracite and 2-4 wt% CaCO3 - is recommended to achieve a balance among safety, combustion efficiency, and economic viability.
While premodern hominins utilized highland landscapes in East Africa as early as 2 million years ago, the processes by which hominins first utilized southern Africa’s highest mountains - the Maloti-Drakensberg - are relatively unknown. This paper introduces a new site in these southern mountains, Likonong Shelter. The results of our luminescence dating and excavations establish Likonong as the earliest-known site in Highland Lesotho and demonstrate a hominin presence in the Maloti-Drakensberg Mountains as early as 242,000 years ago. This first suite of archaeological, geochemical and sedimentological data from Likonong charts the progression of hominin-highland engagements from ephemeral encounters during interglacial Marine Isotope Stage (MIS) 7 to the intentional exploitation of the landscape under a broader range of environmental conditions. We suggest that social and demographic transformations during the Middle Pleistocene may have triggered highland occupation pulses and potentially cumulative cultural evolution, generating adaptive strategies that allowed for more sustained highland occupation during the Middle to Late Pleistocene transition (occupations dating from late MIS 6 to MIS 5c). Importantly, the stresses placed on foragers by sub-freezing temperatures, snow, and extreme resource stress would have required more innovative and collaborative solutions requiring both behavioral flexibility and the attainment of certain demographic preconditions.
This review synthesizes the key scientific discussions and clinical insights presented at the Core Conference of GlomCon Hawaii 2025, held in Oahu, Hawaii, from September 23–25, 2025.
Accurate lesion quantification represents a critical component of precision diagnostics and targeted therapeutic strategies, yet current methodologies face challenges when confronted with the diverse contextual and complicated structures inherent in visible-light medical imaging, including semantic ambiguity, noise interference, and geometric complexity, which collectively hinder segmentation accuracy. Targeting these challenges, we proposes the Multi-Aspect Large Vision Model (MasLVM), a foundational model for optical medical imaging that achieves comprehensive feature fusion across Tri-Path fusion. The Semantic Context Encoder (SCE) integrates a pre-trained large vision model with global semantic embeddings to improve contextual abstraction and mitigate semantic ambiguities. The Spectral Spline Encoder (SSE), incorporating the Multi-Frequency Feature Modulator (MFFM) and Kolmogorov-Arnold Networks (KAN) Channel Attention, transitions image representations into the frequency domain to selectively attenuate noise while preserving essential structural features. The Hierarchical Deformable Morphometry Encoder (HDME) employs deformable convolutions and multi-scale encoding to capture heterogeneous geometric structures dynamically. The outputs from these branches are synthesized through the Multi-Attention KAN Decoder, which employs KAN multiple self-attention and iterative attentional fusion to select and enhance semantic, spectral, and morphological critical domain features adaptively. Extensive experiments across six widely recognized datasets demonstrate that MasLVM achieves convincing performance compared with multiple previous state-of-the-art (SoTA) methods, and potential utility in adapting to diverse requirements of visible-light medical imaging tasks under constrained conditions. The code and model weights can be directly used for medical task deployment or fine-tuning, and are publicly available at the following link: https://github.com/IMOP-lab/MasLVM-Pytorch.
