Falls in older adults remain a major clinical concern. Although the Enhanced Paper Grip Test (EPGT), provides an objective measure of lower limb strength and has shown promise in controlled settings, its use in routine podiatric practice has not been described. To audit the implementation of the EPGT in routine private podiatric practice and to explore whether EPGT measures are associated with self-reported recent falls in older adults. The EPGT was introduced on a trial basis in three independent private UK podiatry clinics for six months. People (aged ≥60 y) were tested when the treating podiatrist considered the assessment clinically appropriate. Anonymised routine clinical data were extracted retrospectively at the end of the audit period, including EPGT outcomes, age, sex, and self-reported history of an unexplained fall in the previous twelve months. Seventy-eight participants (mean age 72 ± 7 years) were assessed; 17 reported a fall. Median EPGT force was statistically significantly lower in fallers (11.9 N) than in non-fallers (21.4 N), U= 283,z = -3.554,p < .0005. Differences remained significant after adjustment for age and sex and were robust to outlier exclusion. Hallux-related pathology was not associated with EPGT outcomes. In this pragmatic private practice audit, the EPGT could be incorporated into routine podiatric assessment and lower EPGT values were associated with self-reported recent falls. These findings support the EPGT as a potentially useful adjunct to falls risk screening that could be offered as part of routine podiatry care. Prospective studies are needed before predictive use or clinical thresholds can be established.
In the current study, N-doped apple tree branch-based biochar-supported Co/CoO catalyst (Co/CoO@NBCs) was prepared using an impregnation-pyrolysis synergistic strategy, which addressed the issues of easy aggregation, poor stability, and low specific surface area of the Co/CoO catalyst. The characterization results show that Co/CoO@NBCs has a large specific surface area (1871.02 m2/g), a developed pore structure (0.99 cm3/g), and Co exists in the form of Co atoms and CoO. Under conditions of 0.25 mM peroxymonosulfate (PMS), 10 mg of catalyst, and room temperature, the degradation rate of norfloxacin (NOR, 30 mg/L) reached equilibrium within 70 min, with a removal rate of 99.06%. Besides the strong alkali condition (pH = 11) and the presence of HCO3-, the removal rate of NOR by Co/CoO@NBCs in other pH ranges and in anionic environments can still exceed 73%. Furthermore, analysis of the degradation mechanism shows that the active sites of Co0 and Co2+ can activate PMS to produce reactive oxygen species, with O2-• and 1O2 serving as the main active substances involved in NOR degradation. Additionally, tests with real water systems and cycling experiments further demonstrate that Co/CoO@NBCs have practical application potential. This work offers new insights into designing heterogeneous catalysts with multiple active sites for wastewater treatment.
The vertically aligned carbon nanotube (VACNT) array, characterized by a vertical arrangement of carbon nanotubes at the bottom and a chaotic crown-like structure at the top, is commonly employed in optical ultrasound emitters due to its high light absorptance and anisotropic thermal conductivity. However, the excessively dense carbon nanotubes at the bottom hinder the incorporation of elastic materials, which necessitates regulation of its morphological structure. To validate the proposed hypothesis, a series of VACNT arrays were synthesized using the chemical vapor deposition technique, which enabled precise control over the morphological properties of the arrays by adjusting the growth duration. Experimental results reveal that an 8 µm thick sample achieves the optimal performance balance, yielding a peak sound pressure of approximately 9.6 MPa, a -6 dB bandwidth of 21.2 MHz, and an energy conversion efficiency of 0.39% under 20mJ laser excitation. Microscopic structural analysis reveals that at this thickness, the carbon nanotube array achieves a vertically aligned yet relatively sparse configuration at its base, which effectively facilitates efficient directional heat transfer while ensuring sufficient infiltration of the elastomer. These findings provide critical insights into the design, fabrication, and performance optimization of high-efficiency VACNT-based photoacoustic transmitters.
This review examines the use of polymer solutions for in-situ subsurface remediation, with a focus on their rheological behavior and implications for contaminant removal. The in-situ remediation of subsurface contamination is often constrained by aquifer heterogeneity, preferential flow, and limited reagent contact with trapped contaminants. Polymer solutions, particularly shear-thinning biopolymers such as xanthan gum (XG), have emerged as promising tools to overcome these challenges. Originally adapted from enhanced oil recovery applications, their unique rheological properties allow high injectivity near the well while promoting mobility control farther into the formation. This enables more stable displacement fronts, suppression of viscous fingering, and enhanced crossflow into low-permeability zones, thereby improving Non-Aqueous Phase Liquids (NAPLs) recovery while also enhancing contaminant removal and amendment delivery in low-permeability regions of heterogeneous media. Beyond direct displacement, polymers may act as carriers for a wide range of remedial amendments, including oxidants, reducers, electron donors, surfactants, and nanoparticles, improving their placement, persistence, and effectiveness. Yield-stress and densified formulations further expand applications by blocking preferential pathways or counteracting buoyancy forces in gravity-dominated systems. Field demonstrations confirm these benefits: Polymer-amended oxidants and electron donors have produced larger swept volumes, more homogeneous propagation, and longer remanence than water-based solutions, with electrical resistivity tomography and coring providing direct evidence of improved distribution and contact. Industrial-scale applications have also shown that formulation and injectivity must be carefully balanced to avoid excessive pressures, fracturing, or reagent incompatibility. Continued integration of laboratory rheology, numerical models, and field validation will be essential to fully realize polymers as multifunctional technologies for contaminant displacement, amendment delivery, and unwanted flow blocking. With growing field evidence, polymer solutions are poised to become central to the design of predictable, durable, and site-specific remediation strategies.
Decades of research have uncovered the complex signaling network downstream of the opioid receptors and suggested how this signaling could be modulated to improve opioid therapy. In our study, we have found that heat shock protein 90 (Hsp90) regulates downstream opioid signaling oppositely in the brain vs the spinal cord. In the spinal cord, we have found that Hsp90 inhibition enables antinociceptive signaling and disables pronociceptive signaling to enhance opioid pain relief and reduce side effects. We have now extended this study to analyze the contribution of protein kinase C (PKC) to the opioid signaling cascade. We used the Hsp90 inhibitor 17-AAG along with a PKC activator or inhibitor (Go6983) delivered into the spinal cords of male and female CD-1 mice to show that pan-PKC signaling contributes to the enhanced opioid antinociception observed in tail flick and postsurgical pain models. We used Western blot and immunohistochemistry to observe increased pan-PKC phosphorylation across calcitonin gene-related peptide (CGRP) and IB4 nociceptors in the spinal dorsal horn. We then used selective siRNA to identify PKCβ as the active isoform and further found PKCβ to be selectively activated in CGRP neurons by Hsp90 inhibition and morphine combined. Finally, we used cell-type-selective Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) to knock down PKCβ in CGRP neurons and showed that this specific isoform in these specific cells was necessary for enhanced opioid antinociception after Hsp90 inhibition.Together, these studies further uncover the novel Hsp90-regulated opioid signaling cascade and suggest how Hsp90 inhibitors could be used to improve opioid therapy by increasing analgesic efficacy and decreasing side effects.
Non-small-cell lung cancer (NSCLC) mortality remains high because mutation-specific therapies target only small patient subsets and inevitably encounter drug resistance. In this regard, multipathway-assisted mesenchymal stem cell-derived nanovesicles (stemsomes) offer a promising universal platform for overcoming the unmet needs of lung cancer therapeutics, particularly for patients lacking identifiable oncogenic drivers. This study introduces a novel strategy in which dexamethasone, a compound conventionally used as an anti-inflammatory drug, is repurposed to enhance the tumor-targeting capabilities of stemsomes. Nanoparticles engineered by fusing these dexamethasone-primed stemsomes with liposomes exhibit markedly improved therapeutic effects. According to transcriptomic and siRNA-mediated knockdown experiments, this enhanced targeting of tumor cells is driven by the dexamethasone-induced upregulation of key cell adhesion proteins, specifically ephrin type-A receptor 2 and neurogenic locus notch homolog protein 3. Furthermore, a comprehensive map of potential adhesion interactions and computational simulations suggest a multivalent interaction network between the surfaces of dexamethasone-primed stemsomes and NSCLC H1975 cells. These findings indicate the discovery of a highly translational and mutation-independent strategy that represents a promising novel mechanism for engineering vesicle surfaces for NSCLC therapy.
In computational pathology, Hematoxylin and Eosin (H&E) staining offers a cost-effective solution for tissue analysis, while Immunohistochemistry (IHC) delivers specific biomarker expression at substantially higher cost and operational complexity. Existing H&E-to-IHC translation methods predominantly operate at the pixel level, often overlooking the preservation of high-level semantic features required by modern multi-instance learning frameworks. To bridge this gap, we present FeatStainDiff, a diffusion-based model that performs direct feature-level transformation between staining modalities. Our framework incorporates two novel components: a Contrastive Semantic Bridging mechanism that ensures diagnostic semantics are preserved during cross-modal translation, and a Frequency-domain Mixture of Experts module that adaptively handles distribution shifts through spectral processing. This design enables the generation of high-fidelity and pathologically consistent IHC features directly from H&E inputs. Through extensive evaluation on two virtual staining datasets and two whole-slide image classification benchmarks, we demonstrate that FeatStainDiff consistently surpasses existing approaches. The method achieves significant improvements in feature similarity metrics, while downstream classification tasks benefit from markedly enhanced performance. FeatStainDiff provides an effective and practical pathway for computational biomarker prediction, with promising potential to expand access to specialized staining analysis in resource-limited clinical environments. Code will be made publicly available upon publication.
Protein tyrosine phosphatase nonreceptor type 22 (PTPN22) is a key negative regulator of T cell activation, acting with C-terminal Src kinase (Csk) to suppress early T cell receptor (TCR) signaling and maintain immune tolerance. Given that the autoimmune disease-associated R620W variant alters T cell responses, we investigated the effects of PTPN22 on T cell activation. We identified a role for PTPN22 in modulating cytoskeletal dynamics at the immunological synapse in Jurkat cells through its interaction with proline-serine-threonine phosphatase-interacting protein 1 (PSTPIP1), a cytoskeletal adaptor protein that recruits actin nucleation-promoting factors, including WASp, to the TCR. PTPN22 deficiency or inhibition disrupted Arp2/3-dependent actin remodeling, leading to excessive central F-actin foci, PSTPIP1 mislocalization, and enhanced Ca2+ signaling, especially under low-affinity stimulation of the TCR. Super-resolution DNA-PAINT analysis revealed that loss of PTPN22 promoted aberrant PSTPIP1-TCR nanoscale colocalization and increased TCR clustering. These findings uncover a PTPN22-PSTPIP1 signaling axis that is critical for regulating cytoskeletal remodeling and receptor organization, providing insights into T cell hyperactivation that may be relevant to autoimmune disease.
The aim of this study was to investigate the effects of ultrasound-assisted tumbling (100, 300, 500, and 700 W) with different treatment times (30, 60, 90, 120, 150, and 180 min) on the quality, myofibrillar protein structure, and protein oxidation of beef (knuckles, 48 h postmortem). Differences in physicochemical properties were assessed, including tumbling yield, protein content in the brine, cooking loss, color, texture, and moisture distribution. Additionally, protein structural and oxidative characteristics were analyzed by measuring carbonyl groups, sulfhydryl groups, and secondary structural conformations. Results indicated that ultrasound-assisted treatment increased both tumbling yield and brine protein content (P < 0.05). Compared with single tumbling (conventional tumbling without ultrasound treatment), ultrasound-assisted tumbling increased the L* value and myofibrillar fragmentation index, while reducing the a* value, hardness, and shear force (P < 0.05). Magnetic resonance imaging, low-field nuclear magnetic resonance, and cooking loss results confirmed that the ultrasound process raised the bound water content and enhanced water-holding capacity of beef (P < 0.05). Moreover, increased ultrasonic power elevated protein carbonyl content and reduced total sulfhydryl content, thus promoting protein oxidation. Conformational analyses revealed that ultrasound treatment reduced protein α-helix content while elevating β-sheet content. In summary, ultrasound-assisted tumbling improved cured beef quality by regulating water migration and modifying protein and myofibrillar structures.
Lacto-N-biose I (LNB), a core structural unit of human milk oligosaccharides, was usually fermented by microbial cell factories because its biosynthesis involved the supplies of ATP and UDP-sugars. Here we designed an in vitro ATP- and UDP-sugar-free enzymatic pathway for the biosynthesis of LNB from lactose and N-acetylglucosamine (GlcNAc). Thermoclostridium caenicola cellobiose phosphorylase was discovered to have a very high promiscuous activity of lactose phosphorylase. Its lactose phosphorylase activity was enhanced greatly by directed evolution, yielding the variant M2 (N654A/Y501A) having 1.56-fold higher lactose activity and doubled lactose/cellobiose specificity. The coenzyme-free three-enzyme molecular machine containing lactose phosphorylase M2, lacto-N-biose phosphorylase and polyphosphate glucokinase produced up to 115 g/L LNB from lactose and GlcNAc and exhibited the highest volumetric productivity of 14.4 g/L/h. This coenzyme-free multienzyme molecular machine could provide a cost-competitive platform for in vitro biomanufacturing of LNB.
Perovskite solar cells (PSCs) offer exceptional tunability of optoelectronic properties, enabling wide-band-gap absorbers that are highly attractive for semitransparent devices in building-integrated photovoltaics (BIPV). However, challenges associated with stability, scalability, and materials' cost continue to limit their practical deployment, highlighting the pivotal role of hole transport materials (HTMs) in achieving high efficiency and durable device operation. Herein, we report the rational design and synthesis of three novel small-molecule HTMs based on phenothiazine-triarylamine cores, prepared via concise synthetic routes with moderate-to-high yields. The electron-rich, nonplanar phenothiazine scaffold enables suppressed aggregation and favorable energy-level alignment, rendering these materials particularly suitable for wide-band-gap and semitransparent PSCs. When implemented in FAPbBr3-based semitransparent devices, two candidates (SM1 and SM2) achieve power conversion efficiencies comparable to those of the state-of-the-art poly(triarylamine) (PTAA) (PCE = 6.26% and 6.09% for SM1 and SM2, respectively, vs 6.39% for PTAA). Notably, their enhanced optical transparency leads to comparable light-utilization efficiency (LUE) (4.05 and 3.99 for SM1 and SM2, respectively, vs 4.07 for PTAA), with outstanding and superior bifaciality factors (84% and 82% for SM1 and SM2, respectively, vs 81% for PTAA), providing a distinct advantage beyond conventional opaque-PV efficiency metrics. These findings position phenothiazine-based HTMs as promising, cost-effective alternatives to PTAA for scalable semitransparent perovskite solar cells.
Oxytocin modulates social information processing by altering excitatory-inhibitory balance at the microcircuit level, but how such local modulation gives rise to selective processing at the level of distributed brain systems remains unclear. Here, we investigated the effects of oxytocin on large-scale neurodynamics across cortico-limbic network in the mouse brain using multisite local field potential recordings. Oxytocin selectively enhanced neural responses to infant calls in the auditory cortex (AC) and medial prefrontal cortex (mPFC). These enhancements occurred while baseline activity was reduced, indicating increased signal-to-noise ratio rather than a global increase in excitability. During auditory steady-state responses (ASSRs), oxytocin increased prefrontal phase coherence without altering ASSR power. During rest, oxytocin induced a transient, broadband reduction in spontaneous spectral power across regions. Despite this reduction in activity, analyses of interregional interactions revealed a selective increase in low-theta phase coupling and directional connectivity of AC→mPFC. Session-level analyses showed that stronger bottom-up AC→mPFC coupling was associated with lower prefrontal power, consistent with a gating or disinhibitory network regime favoring sensory-to-prefrontal information transfer. Multivariate analyses showed that oxytocin/saline conditions were reliably discriminable using supervised classification models, with specific contributions from spectral power, phase-locking, and Granger-causal connectivity features. Conversely, unsupervised dimensionality reduction did not identify a distinct low-dimensional manifold separating conditions, although a modest shift in the centroid of neural state space was observed. Together, these results indicate that oxytocin reduces background neural activity while selectively enhancing sensory-prefrontal network interactions, providing a systems-level account linking local inhibitory modulation to selective processing of socially salient infant cues.
Phosphoglycerate dehydrogenase (PHGDH), a key regulator in the serine biosynthesis pathway, is aberrantly expressed in various cancers, making it an attractive therapeutic target. In this study, we designed and synthesized a series of PHGDH inhibitors using a structure-based approach. Among these, compounds 43 (GDD-260) and 47 (GDD-261) exhibited superior enzymatic inhibition with IC50 values of 0.091 ± 0.013 μM and 0.061 ± 0.004 μM, respectively. Both compounds effectively suppressed de novo serine biosynthesis and showed antitumor activity in PHGDH-overexpressing MDA-MB-468 and PC9 cells. Notably, compounds 43 and 47 also demonstrated antiproliferative effects against erlotinib-resistant PC9 and HCC827 cell lines, exhibiting synergistic effects when combined with erlotinib. Compound 47 showed enhanced antitumor efficacy in erlotinib-resistant PC9 xenograft models in combination with erlotinib. The X-ray crystallographic analysis revealed the binding mode of 43 within the PHGDH active site. These findings provide a foundation for developing PHGDH-targeted anticancer therapies.
The transition toward value-based care in the United States has introduced episode-based payment models that increasingly tie physician reimbursement to longitudinal costs and standardized outcomes. The Centers for Medicare & Medicaid Services (CMS) Ambulatory Specialty Model (ASM), targeting chronic low back pain (cLBP), represents a pivotal extension of this framework into interventional pain management. While intended to reduce low-value utilization, such models risk redefining clinical success in ways that may not align with the heterogeneous and biopsychosocial nature of chronic pain. This perspective examines the potential for outcome-driven reimbursement to incentivize risk selection, marginalize clinically meaningful but non-durable functional gains, and exacerbate existing health disparities. Based on available literature, we propose a "Value Plus" framework incorporating enhanced risk adjustment and patient-centered composite outcomes to better align economic incentives with the realities of chronic pain care. In conclusion, in the field of interventional pain management, withdrawal of multiple conflicting models (ASM, WISER, etc.), and addition of a payment model including a value plus framework would better align incentives with the realities of delivering care to chronic pain patients.
Accumulating preclinical evidence has highlighted the importance of cerebrospinal fluid (CSF) compartmentalization and transport. However, detailed structural characterization in humans remains challenging. This study utilized contrast-enhanced T2-fluid-attenuated inversion recovery imaging on 3-Tesla magnetic resonance imaging (MRI) in a cohort of 477 patients, primarily with reversible cerebral vasoconstriction syndrome (RCVS), to provide a noninvasive in vivo diseased model to delineate distinct subarachnoid compartmentalization and potential leptomeningeal arteriovenous perivascular shunting. We characterized a homogeneous CSF milieu within the ensheathed periarterial glymphatic space. This environment, structurally defined by the perivascular membrane, exhibited uniform tracer intensities across both proximal and distal arterial segments (p = 0.118) on both static and dynamic models, opposed to the heterogeneous CSF appearance observed outside the perivascular membrane. We also observed nodal tracer enrichment in specific locations of the leptomeningeal perivenous space (PVeS) across initial and follow-up MRI. Furthermore, the periarterial tracer enrichment intensities matched those in these nodal portions of the PVeS but significantly exceeded those in the non-nodal portions (p < 0.0001). A dynamic MRI subgroup analysis further revealed that the periarterial tracer kinetics were nearly identical to those of these PVeS nodes. Notably, the nodal-non-nodal gradient of the tracer intensity was significantly amplified in participants exhibiting overt periarterial tracer leakage (p = 0.0006). Although we could not directly visualize the arteriovenous perivascular shunting demonstrated in animal models, our findings may be supportive of potential human periarterial and meningeal border pathways. By establishing a diseased model-based imaging framework to characterize these glymphatic microstructures noninvasively, our results offer a preliminary basis for understanding the perivascular CSF environment and a hypothesized periarterial and meningeal border pathways in living humans.
The firmest evidence in favor of models that posit early high-level influences of cognition on perception comes from electroencephalography (EEG). Enhanced early, preattentive processing of light and dark blue feature changes compared to light and dark green changes was reported in Greek speakers, who have two basic terms for "blue" (ghalazio/ble). In the present three-experiment study, we systematically reevaluate this evidence and test an alternative model that the early difference waves in the human EEG instead mainly reflect contrast adaptation phenomena. We use the same classical oddball paradigm presenting alternating standards and deviants that systematically differ in color and/or luminance and chromatic contrast. We then calculate the visual mismatch negativity (vMMN), a putative index of preattentive feature change processing and predictive coding derived from EEG data, by subtracting the activity elicited by standards from that elicited by task-irrelevant deviants. Our experiments demonstrate the following: 1) vMMN is driven by contrast adaptation, being observable only in the presence of contrast differences between the stimuli and not reliably observed for categorically different hues equated in contrast; 2) there is no reliable difference between green- and blue-related difference waves in speakers (Russian) with two basic blue color categories, the difference waves, again, being driven by contrast rather than their categorical content. Our findings are highly significant for the debate concerning the interface between perception and cognition, as the absence of early categorical effects speaks against models that predict Whorfian-type preattentive cognitive influences on perception.
Element doping has attracted extensive attention as a strategy to regulate the electronic structure of catalysts. In this study, Ni-doped Bi2MoO6 nanosheets were uniformly anchored onto rod-like In2O3 to furnish substantial accessible reactive sites and close interfacial contact. Ni doping enabled a precise modulation for the band configuration of Bi2MoO6, imparting a Z-scheme heterojunction between Ni-doped Bi2MoO6 and In2O3. The Ni-doped Bi2MoO6/In2O3 Z-scheme heterojunction exhibits excellent photocatalytic performance in the treatment of Cr(VI) and emerging organic contaminants. In the Cr(VI)/bisphenol A coexistence system, Ni-doped Bi2MoO6/In2O3 not only achieved high reduction efficiency for Cr(VI), but also enhanced the degradation rate of bisphenol A by a factor of 4.7 compared to the single-pollutant system. The removal efficiencies could surpass many related reported studies. The product toxicity after bisphenol A degradation in a mixed Cr(VI)/bisphenol A system was analyzed by the total organic carbon content, high-performance liquid chromatography-tandem mass spectrometry, the toxicity estimation software tool, and microbial activity, manifesting that their impact on the environment has been greatly reduced. Furthermore, the catalyst maintains stable and consistent performance in multiple binary systems, including Cr(VI)/tetracycline and Cr(VI)/ciprofloxacin. This work provides insights for the design of element-doped heterojunction catalysts, which demonstrate high performance, durability, and environmental compatibility in the treatment of complex wastewater.
Radiofrequency ablation (RFA) is becoming a standard treatment for early-stage hepatocellular carcinoma (HCC). However, high rates of postoperative recurrence, particularly following insufficient RFA (iRFA), remains a clinical obstacle. iRFA contributes to the formation of an immunosuppressive tumor microenvironment and facilitates tumor progression, though the molecular mechanisms driving these processes are not fully elucidated. This study seeks to comprehensively characterize the spatial architecture and cellular interactions within iRFA HCC, and to identify key transcriptional mechanisms through which sublethal heat stress promotes immune evasion and recurrence. We integrated high-definition spatial transcriptomics (Visium HD) with single-cell RNA sequencing data to map the cellular heterogeneity and communication networks in samples from 6 HCC patients who underwent surgical resection after iRFA. Functional validation was performed using in vitro heat stress models, co-cultured assays, and orthotopic mouse models. We provided a high-resolution spatial atlas of iRFA HCC and the ablation zone, identifying a subpopulation of tumor cells exhibiting activation of the transcription factor CEBPD in response to sublethal heat stress. CEBPD directly bound to the promoter of CXCL2 and drove its expression. Up-regulation of CXCL2 diminished immune checkpoint inhibitors response in iRFA HCC through promoting SPP1+ TAM accumulation via the CXCL2/STAT3/SPP1 axis, and enhanced tumor cell invasion through autocrine signaling. Our study provides a comprehensive spatial characterization of HCC following iRFA, identifies the heat stress-CEBPD-CXCL2 axis as a key driver of tumor progression post-RFA, and demonstrating the therapeutic potential of CXCR2 inhibition for preventing tumor recurrence following RFA.
Gas-driven element redistribution, characterized by the preferential enrichment of one element at the surface relative to the bulk, is frequently observed in multicomponent alloys. Using L10-ordered PtNi as a model system, we reveal that gas pressure plays a critical role in governing adsorption-driven surface composition during annealing in reducing gases: low pressure favors Pt surface segregation, while high pressure facilitates Ni surface enrichment. In this study, we developed a high-pressure nitriding (HPN) strategy that modulates the surface structure and composition of PtNi catalysts. The resulting HPN-PtNi exhibits enhanced performance and durability in membrane electrode assemblies for heavy-duty fuel cell applications, maintaining a high current density of 1.19 A cm-2 at 0.7 V after 90,000 voltage cycles. Through a combination of experimental and theoretical analyses, we reveal that the HPN process forms additional stabilizing Ni-N bonds and induces elemental redistribution with Ni surface enrichment and a Ni-deficient Pt subsurface. These modifications alter the atomic coordination environment of the ordered PtNi phase. This work presents a generalizable strategy to design robust and high-performing Pt-based catalysts by controlling gas-pressure-driven elemental redistribution and dopant incorporation.
Accurate classification of sea turtle species is crucial for ecological monitoring and conservation, yet traditional visual classification methods remain limited by underwater imaging challenges such as occlusions, poor lighting, and background noise. To address these limitations, we propose an enhanced deep learning-based classification framework that integrates both color and structural features to improve the robustness of species recognition in complex marine environments. Building upon the ResNet-50 backbone, we introduce a four-channel input tensor comprising RGB data and Sobel-filtered edge maps, capturing both semantic and morphological information. Two novel fusion modules, LiteAFNet and AlphaBlendNet, are designed to integrate these features effectively. LiteAFNet leverages a lightweight attention mechanism to highlight discriminative regions, while AlphaBlendNet adaptively balances RGB and edge cues based on spatial context. Experimental results demonstrate significant improvements in classification performance across all evaluation metrics. Specifically, AlphaBlendNet achieves the highest precision (0.84), recall (0.88), F1-score (0.86), and mean average precision (mAP) of 87.2%, outperforming both the baseline fusion and LiteAFNet configurations. These results indicate that integrating color histograms with structural edge features enhances the model's ability to distinguish between species with similar visual traits. This framework offers a scalable, accurate, and automated solution for underwater species classification and holds potential for broader application in marine biodiversity monitoring.