Search Results (9,302)

Fast charging is critical to the wider adoption of electric vehicles, but simply increasing the charging current intensifies polarization and degradation, creating an inherent conflict between the charging speed and battery lifetime. Multi-stage charging can alleviate this conflict by redistributing the current throughout charging, although its performance depends jointly on the protocol structure and stage parameters. To address this problem, this work proposes a hierarchical multi-objective optimization method that coordinates these two design levels. A reduced-order electrochemical–thermal–aging model, developed and validated using systematic degradation experiments, is employed to predict electrothermal responses and capacity loss throughout the battery lifetime. For each candidate stage number, the charging rates, switching voltages, and exit-current ratios are jointly optimized, while the resulting Pareto performance and implementation complexity are compared at the structure level. The results showed that a three-stage structure captures most of the attainable performance gains without unnecessary control complexity. Full-lifetime cycling experiments demonstrated that the representative protocols achieved different balances between charging speeds and cycle lives. The 4C-referenced protocols extended cycle lives by 25.2–35.9% with only 2.7–4.1% longer initial charging times. The 3C-referenced protocols shortened initial charging times by 5.6–8.6%, while their cycle lives remained broadly comparable to 3C CCCV, ranging from a 9.5% decrease to a 6.9% increase. Multi-level degradation analyses further associated the lifetime improvements with a slower accumulation of anode-related capacity loss and interfacial deposits, together with reduced impedance growth. Full article

Drug–target affinity (DTA) prediction plays an important role in computational drug discovery; however, its application to emerging infectious diseases such as Ebola remains challenging because of the limited availability of experimentally measured affinity data. To address this low-resource setting, we propose a retrieval-guided transfer-learning framework that leverages BindingDB interactions to improve Ebola DTA prediction. The framework uses a two-stage strategy. In Stage I, source interactions are selected using random sampling, compound-similarity retrieval, protein-similarity retrieval, or hybrid compound–protein retrieval at source-data budgets of 50,000 and 300,000 interactions and combined with Ebola training data to learn transferable representations. In Stage II, the pretrained compound and protein encoders are frozen, while the prediction layers are adapted to the Ebola domain. The framework was implemented with DeepDTA and GraphDTA and evaluated across five random seeds using a scaffold-based split, with conventional machine-learning models and single-stage deep learning as baselines. The best overall configuration, GraphDTA with protein-similarity-guided retrieval at the 50,000-interaction budget, achieved RMSE $=0.5498\pm 0.1073$, $R^2=0.8809\pm 0.0452$, and Pearson $=0.9395\pm 0.0237$, outperforming the strongest conventional machine-learning model (Extra Trees, RMSE $=0.6065\pm 0.0642$) and single-stage GraphDTA (RMSE $=0.6533\pm 0.0752$). Across both source-data budgets and both DTA backbones, all targeted retrieval configurations achieved lower mean RMSE than their corresponding random-retrieval configurations. At the 50,000-interaction budget, matched seed-wise analysis further showed consistent improvements for protein-guided retrieval across all five seeds for both backbones. Retrieval characterization showed stronger target-domain similarity and substantially lower cross-strategy overlap at 50,000 than at 300,000 interactions, while post hoc sequence analysis independently confirmed enrichment of sequence-level relatedness with protein-guided retrieval. Increasing the source-data budget from 50,000 to 300,000 did not uniformly improve predictive performance, indicating that source-data relevance and retrieval selectivity should be considered jointly with source-data quantity. Finally, virtual-screening and approved-drug repurposing case studies across six Ebola virus targets demonstrate the use of the framework for computational prioritization of compound–target hypotheses. Overall, the findings support relevance-guided source-data selection as an effective strategy for transfer learning in low-resource DTA prediction. Full article

Salinity gradient energy, also known as blue energy, is a clean and renewable option for electricity generation with no direct CO${}_2$ emissions. Among the available technologies, pressure-retarded osmosis (PRO) stands out, although its large-scale implementation is not yet economically viable, so predictive models are essential to assess its real potential. In the present work, a multistage PRO system of up to three stages, with one–three hollow-fiber membrane modules (HFMMs) arranged in series per stage, was simulated and the operating conditions were optimized. The model accounts for axial pressure drops, feed concentration and draw dilution along each module, considering two salinity gradients of 29.5 and 59.5 g L${}^{-1}$. The maximum net specific energy generation reached 421.48 Wh m${}^{-3}$ with three stages under the 59.5 g L${}^{-1}$ gradient. Adding HFMMs in series benefited the two-stage system but slightly penalized the three-stage one, and the net specific energy decreased monotonically as the ratio of pressure vessels between the first and the second stage increased. At least three stages are therefore required for a full-scale PRO system to become a net energy producer at moderate salinity gradients, and further energy should be sought through additional staging rather than through longer series. Full article

Endourological stone surgery frequently relies on intraoperative X-ray fluoroscopy for guidewire and access sheath placement, retrograde pyelography, stent positioning, and percutaneous renal access. Although radiation exposure from an uncomplicated procedure is usually limited, recurrent imaging and interventions contribute to cumulative patient exposure, and repeated procedures result in occupational exposure among operating-room personnel. This narrative review examines fluoroscopic workflows in ureteroscopy, retrograde intrarenal surgery, percutaneous nephrolithotomy, and endoscopic combined intrarenal surgery, with particular emphasis on dose metrics, determinants of exposure, and practical optimization. Readily implementable measures consistent with the as-low-as-reasonably-achievable (ALARA) principle include low-dose and pulsed fluoroscopy, reduced pulse rates, collimation, last-image hold, optimized C-arm geometry, protective equipment, dosimetry, feedback, and team training. Evidence also supports fluoroscopy-free or fluoroscopy-minimized ureteroscopy and retrograde intrarenal surgery in selected patients, as well as ultrasound-guided percutaneous nephrolithotomy, including hybrid ultrasound-fluoroscopy workflows. Endoscopically assisted puncture, three-dimensional reconstruction, image fusion, navigation, augmented reality, and computer vision span different stages of clinical maturity; AI-assisted dose optimization remains experimental and lacks endourology-specific clinical validation. Standardized multidimensional reporting of patient and staff dosimetry, evidence-based technology classification, and procedure- and complexity-specific benchmarks are needed. Prospective multicenter studies should evaluate integrated image-guided workflows using clinically relevant outcomes, including radiation dose, stone-free outcomes, complications, operative time, usability, and cost. Full article

Chronic kidney disease (CKD) confers a high burden of atherosclerotic cardiovascular disease (ASCVD) and a characteristic, evolving dyslipidaemia that is incompletely addressed by conventional care. Statin-based therapy remains first-line in most non-dialysis CKD populations, yet many patients fail to reach lipid goals or remain at substantial residual risk. This structured narrative review evaluates injectable lipid-lowering agents across the CKD spectrum, focusing on the pharmacology and clinical utility of therapies targeting proprotein convertase subtilisin/kexin type 9 (PCSK9). The monoclonal antibodies evolocumab and alirocumab are subcutaneous biologics that bind circulating PCSK9, prevent PCSK9-mediated degradation of the low-density lipoprotein (LDL) receptor, and enhance hepatic LDL-cholesterol (LDL-C) clearance. Their catabolic elimination and negligible cytochrome-P450 interaction burden are advantageous in polypharmacy-heavy CKD care. Inclisiran is a hepatocyte-directed, GalNAc-conjugated small interfering RNA that silences PCSK9 messenger RNA, enabling durable LDL-C lowering with twice-yearly maintenance dosing after initiation. Across major programmes, injectable PCSK9 inhibition produces a ~50–60% LDL-C reduction, with consistent efficacy and safety across renal function strata, although outcome data in advanced CKD and dialysis remain limited. Importantly, cardiovascular outcome benefit is established for the monoclonal antibodies evolocumab and alirocumab, whereas inclisiran currently has robust LDL-C-lowering evidence but no completed cardiovascular outcome trial; these agents should not be regarded as having equivalent outcome evidence. Historical dialysis statin trials showed attenuated or neutral cardiovascular benefit, underscoring the biological and trial-design complexity of late-stage kidney disease and the need for dedicated injectable-era trials. We propose a stage-specific framework for integrating injectable therapies into CKD pathways—covering patient selection, sequencing, monitoring and implementation—and outline priorities for CKD-enriched outcome trials, kidney-relevant endpoints, and the integration of emerging lipoprotein(a)-targeted injectables into cardio-renal risk reduction. Full article

This paper presents a low-power 32.768 kHz multi-crystal oscillator implemented in advanced 18 nm CMOS technology and integrated into a general-purpose microcontroller (MCU) environment. The oscillator operates over a wide supply voltage range from 1.6 V to 3.6 V and across a temperature range from $-40{}^{\circ }\mathrm{C}$ to $130{}^{\circ }\mathrm{C}$, meeting industrial requirements. The experimental results demonstrate a current consumption below 200 nA under typical conditions (i.e., 3 V and $25{}^{\circ }\mathrm{C}$), with reliable start-up and stable oscillations using various crystals that cover a broad range of crystal market characteristics. The oscillator supports crystals with equivalent series resistance (ESR) values up to $100\mathrm{k}\mathsf{\Omega }$ and datasheet load capacitances ranging from $4\mathrm{pF}$ to $12.5\mathrm{pF}$ without amplifier reconfiguration. The design is based on a two-stage amplifier architecture that effectively reduces power consumption while improving frequency stability and predictability. The simulations show that the proposed architecture offers greater robustness than the classical Pierce architecture regarding oscillation frequency variations caused by potential crystal ESR fluctuations. These findings are validated by measurements, which exhibit a high correlation between the simulated and measured oscillation impedance behavior. Full article

Decentralized coordination of multi-agent communication networks underpins autonomous UAV swarms, sensor meshes, and cyber-physical systems that must operate without a single point of failure, yet quantitative design rules linking topology degradation to coordination collapse are still lacking. This paper aims to characterize the fault tolerance of a discrete linear consensus protocol that drives mobile agents’ scalar states toward a common value over a time-varying communication graph. The convergence rate and fault tolerance budget of the protocol are explicitly tied to the algebraic connectivity ${\lambda }_2$ of the graph Laplacian, and three failure regimes—random independent failures ($M_1$), sequential staged failures ($M_2$), and targeted attacks on the highest-degree hubs ($M_3$)—are analyzed using a full-factorial Monte-Carlo experiment of 1620 runs implemented in Python with the Mesa framework and NetworkX. Critical failure thresholds ${\phi }_c$ are 0.42–0.51 for $M_1$, 0.38–0.48 for $M_2$, and 0.28–0.38 for $M_3$; a Kruskal–Wallis test (p < 0.001) confirms the ordering ${\phi }_c^{M^3}<{\phi }_c^{M^2}<{\phi }_c^{M^1}$, providing quantitative design rules for fault-tolerant consensus-driven networks. Full article

Background: Autoimmune hepatitis (AIH) is a chronic immune-mediated liver disease that may progress to cirrhosis. This study aimed to develop and externally validate interpretable machine-learning models for the classification of prevalent cirrhosis in patients with AIH. Methods: The development cohort included 55 patients with biopsy-confirmed AIH. Cirrhosis was defined histologically as F4, whereas F0–F3 was classified as non-cirrhosis. Logistic Regression with L2 regularization, Random Forest, and XGBoost were evaluated. The original stratified 70/30 hold-out analysis was retained, and repeated stratified five-fold cross-validation with 20 repeats was additionally performed to assess internal stability. Primary external validation was performed in an independent histology-matched cohort of 42 patients. Models were applied without refitting, recalibration, or threshold optimization. Discrimination, probabilistic accuracy, calibration, and threshold-dependent classification metrics were evaluated. Results: In the original held-out test set, AUROC was 0.900 for Logistic Regression with L2 regularization, 0.830 for Random Forest, and 0.890 for XGBoost. In repeated cross-validation, mean AUROC was 0.878 ± 0.021, 0.914 ± 0.015, and 0.896 ± 0.015, respectively. In the primary external validation cohort, Random Forest showed the highest numerical discrimination (AUROC 0.810; 95% CI, 0.653–0.933), followed by XGBoost (0.728; 95% CI, 0.566–0.878) and Logistic Regression with L2 regularization (0.716; 95% CI, 0.545–0.875). Random Forest also had the lowest Brier score (0.188). However, confidence intervals were wide and overlapping. Elastography stage alone achieved an AUROC of 0.745, and the numerical improvement of the full Random Forest model was not statistically clear. Conclusions: Machine-learning models integrating routinely available clinical, biochemical, immunological, and elastography-related variables showed preliminary external transportability for the classification of prevalent cirrhosis in AIH. However, the small development and validation cohorts, uncertainty in calibration, and lack of a clearly demonstrated incremental advantage over elastography alone indicate that larger prospective multicenter studies are required before clinical implementation. Full article

Wastewater treatment systems are under increasing pressure to improve nitrogen removal while reducing carbon emissions, yet the deployment of anaerobic ammonium oxidation (anammox) remains constrained by uncertainty about technical readiness, operational robustness, nitrous oxide (N₂O) emissions, life-cycle carbon performance, monitoring capacity, and transferability across wastewater contexts. This study uses dynamic topic modelling and trend assessment of 998 publications from 2001 to 2025 to synthesize deployment-relevant evidence for anammox-based wastewater treatment. The results indicate that the field has shifted from reactor start-up and process-parameter optimization toward microbial regulation, mainstream process integration, coupled nitrogen-removal strategies, and intelligent control. Building on these topic-evolution patterns and reported engineering evidence, this study provides an evidence-based qualitative appraisal of deployment-readiness signals and evidence gaps, distinguishing comparatively mature side-stream applications from mainstream systems that still require monitored demonstrations, transparent N₂O accounting, life-cycle assessment, and locally validated operating data. The study argues that anammox should be evaluated as a technology with potential but conditional carbon-saving benefits: its potential carbon-saving benefits depend on operational evidence specific to each application stage, carbon-accounting credibility, and implementation capacity, rather than assuming that research activity alone justifies broad deployment. Full article

To support the early identification of potential drowning-risk states and improve rescue response efficiency, this paper proposes a UAV-oriented dual-scale detection-pose cascade for frame-level drowning-risk recognition. The framework first performs high-recall preliminary detection on wide-field input images to identify potential drowning targets. The detected target regions are subsequently extracted and resized to construct localized inputs for the second-stage pose-based verification. This software-based target-region refinement simulates the localized high-resolution observation that could be provided by a telephoto camera in a future physical dual-camera UAV implementation. By focusing subsequent analysis on the localized target regions, the second-stage pose model can exploit finer-scale human structural information for drowning-risk state verification, thereby providing decision support for potential drowning detection. To address the challenges of small target scales and severe background interference in wide-field images, a lightweight YOLOv8n-based detection model is developed. An enhanced edge-feature-guided residual convolutional block attention module (EGRCBAM) is introduced, together with a recall-oriented FPIoU loss function designed for hard sample optimization, improving the recall of the drowning category by 17%. For localized target verification, an enhanced YOLOv8n-Pose model is constructed by incorporating a coordinate-aware pose head, spatial attention mechanism, and skeletal structure constraints to enhance human-region localization and pose-based drowning-versus-swimming recognition. The model improves Box mAP@0.5 from 0.768 to 0.816. Comparative experiments on the self-collected dataset demonstrate the effectiveness of the proposed detection and pose-based verification framework. The proposed framework provides a lightweight vision-based solution or UAV-oriented frame-level drowning-risk recognition and offers a potential algorithmic basis for future integration with physical dual-camera UAV platforms. Full article

Background: Slowing the progression of dementia and maintaining mental and social health and well-being are important aspects of health and healthcare for people living with dementia (PLwD). Relationships with family caregivers and unique caregiver insights are also important aspects of the health and well-being of PLwD. Methods: A group-based gaming system was developed to promote health and well-being for PLwD through movement, cognitive exercises, and social interaction. Included were family caregivers of clients living with mild- or moderate-stage impairment in an adult day center selected for the gaming intervention. Prior to implementing the intervention, three separate focus groups were conducted with 21 caregivers of PLwD, divided into each of the three focus groups based on their availability, to explore what it is like to care for someone living with dementia, their loved ones’ mood, and their questions or concerns regarding the gaming intervention. Responses were thematically analyzed. Results: In total, 21 caregivers participated in the focus groups. Six themes were identified: mutuality and strategies for coping with varying emotions; new challenges and successes with connecting and relating; multidimensional challenges with adjusting to decline, help-seeking, and providing care; community, social support, and resilience; technology and community resources as tools to ease caregiving burdens, and interest in digital gaming for PLwD. Conclusions: Our results suggest that group-based digital gaming interventions may be useful to both caregivers and PLwD, by offering meaningful opportunities for social engagement. Full article

Background/Objectives: Determination of blood metabolic profile and hemogram are particularly valuable in animals kept in extensive and semi-intensive farming systems, as within those systems, it is not possible to properly monitor animals. This research investigated changes in the hemogram indicators and blood metabolic profile of Dalmatian Pramenka ewes by comparing different lactation stage and litter size. Methods: This study investigated changes in the hemogram indicators and blood metabolic profile of 33 Dalmatian Pramenka ewes during lactation as influenced by the lactation stage (taken on the 40th, 70th and 100th day of lactation) and the litter size (1 or 2 lambs in litter). Referring to the litter size, 11 ewes had twins, and 22 ewes had one lamb in litter. Results: This study confirmed the significant influence of stage of lactation on the blood metabolic profile (minerals: Ca, P-inorganic, Mg, Fe, metabolites: urea, glucose, triglycerides, ALB, GLOB and A/G ratio, and ALT enzyme) and most of the hemogram indicators (RBC and WBC content of HGB and HCT and portion of monocytes) in the blood of lactating ewes. The influence of litter size on the hemogram and blood metabolic profile was insignificant, yet its only statistically significant effect was determined for urea concentrations in the blood of ewes. Conclusions: When planning quality monitoring procedures of a herd, a blood metabolic profile and hemogram of the animals should be implemented as routine herd-monitoring tools for indigenous sheep breeds raised under extensive production systems. Full article

Background: Percutaneous intramyocardial septal radiofrequency ablation (PIMSRA), also known as the Liwen procedure, is an emerging septal reduction therapy for obstructive hypertrophic cardiomyopathy (OHCM), but standardized early rehabilitation pathways after PIMSRA are lacking. We aimed to develop an enhanced recovery after surgery (ERAS)-informed, risk-stratified stepwise rehabilitation pathway after PIMSRA, describe its documented early inpatient implementation in selected clinically stable patients, report preliminary in-hospital safety observations, and explore its associations with hospitalization outcomes. Methods: This exploratory single-center retrospective cohort study included 102 patients with OHCM who underwent first-time PIMSRA between 1 July 2023 and 3 August 2025 and met the same predefined post-PIMSRA rehabilitation eligibility criteria. Patients were classified according to the care pathway actually received: usual care (n = 53) or early inpatient implementation of the ERAS-informed rehabilitation pathway plus usual care (n = 49). The planned pathway incorporated rehabilitation eligibility screening, risk stratification, staged early mobilization, structured physiological monitoring with predefined stop criteria, multidisciplinary coordination, and patient education. Pathway selection was nonrandomized and reflected patient- or family-related preferences and implementation or logistical factors. Results: The ERAS-informed group had a shorter ICU length of stay than the usual-care group in the primary analysis [20.0 (5.8–26.6) h vs. 23.0 (17.1–45.5) h; Hodges–Lehmann difference, −6.0 h; 95% CI, −16.51 to −0.53; p = 0.022]. Three ICU stays > 100 h occurred in the usual-care group and none in the ERAS-informed group; after excluding these observations, the comparison was attenuated (p = 0.060). The total hospital length of stay was shorter in the ERAS-informed group [8.0 (7.0–11.0) d vs. 9.0 (8.0–13.0) d; difference, −1.0 d; 95% CI, −2.02 to −0.03; p = 0.029], whereas the post-PIMSRA length of stay did not differ significantly (p = 0.058). The admission-to-PIMSRA interval was also not significantly different (p = 0.309). No rehabilitation-related adverse events were recorded in the ERAS-informed group during hospitalization (0/49; exact binomial 95% CI, 0.0–7.3%). Conclusions: Among selected clinically stable patients after PIMSRA, early inpatient implementation of pathway components was achievable, primarily during the period corresponding to planned stages 1–2. No rehabilitation-related adverse events were recorded, providing preliminary patient-level in-hospital safety observations. Fidelity, adherence, completion, and reproducibility of the complete four-stage, risk-stratified pathway were not established. Length-of-stay associations were exploratory and should not be interpreted causally. Full article

There is a growing need for high-efficiency power electronic converters that can effectively convert energy, regulate voltages, and enhance power quality in standalone power generators, as the use of renewable energy sources and battery energy storage devices increases. The quasi-impedance-source inverter (qZSI) has attracted significant interest due to its single-stage buck-boost operation, continuous input current, reduced reliance on passive elements, and increased reliability. In this paper, the control strategy and implementation of the qZSI with battery assistance for standalone photovoltaic energy generation are discussed. To analyze the operational characteristics and design the control strategy of the qZSI, the system equations are linearized around the nominal operating point to develop a small-signal model, from which the direct current (DC) side and alternative current (AC) side transfer functions are derived and used as the basis for controller design. Using the proposed model, hybrid controllers are designed to control the shoot-through duty cycle, maintain DC link voltage stability, and battery charging to achieve stable power generation. Furthermore, the SPWM technique is applied to produce AC power with minimal harmonic content and higher efficiency. Application results show stable dynamic behavior, effective battery energy management, improved voltage regulation, and reduced harmonic distortion in the output waveform. The main contribution is a low-complexity coordinated PI and PR control framework for standalone battery-assisted qZSI operation, experimentally validated under DC- and AC-side disturbances without requiring an additional battery-side power-conversion stage. Full article

Single-photon avalanche diode (SPAD) arrays enable photon-starved applications, including time-of-flight imaging and stellar intensity interferometry. Their astronomical use remains scarcely explored, as the bottleneck is usually not detection but rather acquisition electronics for high-rate event streams. This work presents a modular, event-driven acquisition system for a 64 × 64 SPAD array within the La Palma Quantum Interferometer (LPQI) project, repurposing a LiDAR detector for multi-telescope interferometry. Two stages are used: a Raspberry Pi 5 with a custom board for validation, and an AMD Kria KR260 (Zynq UltraScale+ MPSoC) implementing the Address-Event Representation (AER) handshake in hardware at 100 MHz. The system streams AER events without per-event timestamping; sub-nanosecond time-tagging is left for a future stage based on the White Rabbit protocol. Optical bench tests confirmed spatial detection and localization of photons at a measured throughput of up to ≈124 keps under the highest illumination condition tested, and dark-count-rate characterization showed a rate below 10 Hz for most pixels (median: 1.68 Hz at 27.8 °C); raw per-pixel event-count maps further confirmed, for the first time on this array, the expected 2 × 2 spatial pattern of inter-pixel crosstalk from its shared-cathode pixel groups. The results demonstrate the feasibility of repurposing a LiDAR SPAD sensor and establish an acquisition-electronics baseline to aid the development of a timestamped, multi-telescope system for deployment on five telescopes of the Roque de los Muchachos Observatory. Full article