Search Results (847)

Hydrogen embrittlement poses a critical threat to the durability of metallic components in emerging hydrogen energy infrastructure. Reliable fatigue life assessment in hydrogen-rich environments is, however, severely constrained by the high cost and low throughput of high-pressure testing, resulting in characteristically sparse experimental datasets. Conventional empirical fatigue models struggle to capture hydrogen–mechanical coupling effects, while purely data-driven approaches often suffer from severe overfitting under data-scarce conditions. To address this challenge, this study develops a physics-enhanced learning framework that integrates established fracture mechanics principles with machine learning. Using high-strength GS80A steel as a case study, two complementary strategies are introduced. First, a physically augmented input strategy reformulates raw experimental variables into dimensionless physical descriptors derived from the Basquin and Goodman relations, thereby reducing the complexity of the learning space. Second, a physics-regularized ensemble strategy combines deterministic physical predictions with neural network outputs through a dual-pathway inference scheme, ensuring physically admissible behavior during extrapolation. An automated hyperparameter selection module is further employed to establish a robust data-driven baseline. Comparative evaluation against optimized multi-layer perceptron and support vector regression models demonstrates that the proposed framework significantly improves predictive robustness in small-sample regimes. Specifically, the coefficient of determination (R²) exceeds 0.975, with the root mean square error (RMSE) reduced by approximately 70% compared to the pure data-driven baseline. By systematically embedding mechanistic priors into the learning process, the proposed approach provides a reliable and interpretable tool for fatigue assessment of metallic components operating in hydrogen environments. Full article

Background: There are ongoing attempts to find a reliable, highly sensitive and specific early indicator of myocardial ischemia. Recently, a potential new function for the “non-pregnancy”-related pregnancy-associated plasma protein-A (PAPP-A) protein has been reported in many papers, including that the protein could be used in diagnosing heart conditions. Hence, our study aimed to determine the diagnostic and prognostic significance of PAPP-A protein in individuals diagnosed with non-ST-elevation acute coronary syndromes (NSTE-ACSs). Methods: The study comprised 100 consecutive patients (68 males and 32 females), aged from 42 to 83 years (mean age: 64.2 years). We assessed PAPP-A protein levels, anthropometric measurements, basic laboratory tests, ECG recordings, and coronary angiography for each patient. The participants were subsequently divided into two groups: non-ST-elevation myocardial infarction (NSTEMI, n = 74) or unstable angina (UA, n = 25). Results: The levels of PAPP-A protein in patients with NSTEMI were slightly higher than those in patients with UA, but the difference was not statistically significant (7.93 ± 6.35 mIU/L vs. 6.52 ± 5.45 mIU/L, p = 0.253). Higher PAPP-A protein levels (≥5.83 mIU/L) were associated with a numerically higher, but not statistically significant, risk of NSTEMI (OR = 1.37; 95% CI: 0.56–3.36). After 12 months, there was a significant correlation between the amount of labelled PAPP-A protein and the likelihood of experiencing acute myocardial infarction, cardiovascular death, and the necessity for unplanned coronary angiography. Conclusions: The diagnostic utility of PAPP-A protein in NSTE-ACS is limited, both in the NSTEMI and UA patient groups. However, its measurement can be used to estimate the annual risk for these groups of patients. Full article

Global food security is an increasing challenge due to population growth and the limited availability of natural resources, driving the search for sustainable protein sources. In this context, edible insects such as yellow mealworms (Tenebrio molitor) have emerged as a promising alternative, while probiotics have been widely applied in animal production to enhance growth performance and nutritional quality. This study aimed to evaluate the effects of probiotic supplementation on the growth performance, biomass yield, and nutritional composition of yellow mealworm larvae at laboratory and pilot scales. Three probiotic strains—Bacillus velezensis, Bacillus coagulans, and Pediococcus pentosaceus—were tested at four different dosage levels, using wheat bran and brewer’s spent grain as feed substrates. Larval growth was monitored weekly, and total harvested biomass, proximate composition (dry matter, protein, fat, and ash), amino acid profile, and mineral composition were determined using standardized analytical methods. At the laboratory scale, probiotic supplementation did not result in significant differences in mean larval weight or total biomass (p > 0.05). In contrast, at the pilot scale, significant improvements in larval growth and biomass were observed for specific probiotic treatments, with mean larval weights reaching approximately 140–150 mg and total harvest biomass increases of up to ~15% compared to the control (p < 0.05). Growth curves at both scales followed a sigmoidal pattern with a high correlation between laboratory and pilot experiments (R² = 0.98). Probiotic supplementation did not significantly affect crude protein content, but alterations in fat content, specific amino acid concentrations, and mineral composition were observed at the pilot scale, depending on strain and dosage. Overall, the results demonstrate that probiotic supplementation can enhance yellow mealworm production under pilot-scale conditions, while laboratory-scale trials may not fully capture these effects. These findings highlight the importance of scale when evaluating probiotic strategies and support the potential application of Bacilli-based probiotics to improve the efficiency and nutritional quality of industrial insect production systems. Full article

Background/Objectives: Pilates exercises performed on different apparatuses may elicit distinct acute responses in cardiovascular function and musculoskeletal flexibility, yet comparative data on the immediate effects of reformer (RF), cadillac (CD), and chair (CH) pilates exercises remain unclear. This study aimed to investigate and compare the acute effects of RF, CD, and CH pilates sessions on cardiac autonomic modulation and flexibility in sedentary middle-aged women. Methods: Fifteen participants (mean age: 42.2 ± 1.5 years) completed all three exercise conditions in a randomized crossover design, with sessions separated by at least 72 h. Heart rate variability (HRV) was assessed at pre-exercise, during exercise, and at 10 min intervals up to 40 min post-exercise. Flexibility was measured using standardized sit-and-reach tests at pre-exercise, immediately post-exercise, and 40 min post-exercise. Results: Results revealed a significant condition × time interaction for flexibility (p < 0.010, η²p = 0.207), with the RF session producing greater immediate improvements in flexibility compared to CD (p = 0.030; g = 0.24) and CH sessions (p = 0.030; g = 0.24). Notably, flexibility gains from the RF session were maintained at 40 min post-exercise relative to the CH session (p = 0.035; g = 0.28). In contrast, no significant interactions between condition × time were observed for HRV parameters (p > 0.05). However, the main effect of time was evident across all HRV measures (p < 0.05), indicating post-exercise autonomic modulation independent of apparatus type. Conclusions: These findings suggest that while acute cardiovascular responses may not differ substantially between pilates apparatuses, the RF apparatus may be more effective for immediate flexibility enhancement in sedentary middle-aged women. Practitioners and clinicians may consider selecting apparatus based on specific functional goals, such as improving flexibility, when designing pilates-based interventions for this population. Full article

(1) Background: Falls are a major public health concern in older adults, largely due to age-related declines in proprioception and postural control. Although knee braces are commonly prescribed to enhance joint stability and sensory feedback, their effects on plantar pressure distribution remain unclear; (2) Methods: Thirteen community-dwelling older adults (mean age: 79.6 ± 3.2 years) participated in a repeated-measures study under three conditions: no brace, knee brace A, and knee brace B. Plantar pressure variables were assessed barefoot during quiet standing using a baropodometric platform. Conditions were compared using non-parametric Friedman tests; (3) Results: Significant differences were observed for left foot total surface area (p = 0.041) and left rearfoot surface area (p = 0.020). Compared with no brace, brace A increased plantar contact area, whereas brace B reduced it. No significant differences were found for pressure magnitude, load distribution, or right foot variables; (4) Conclusions: Knee braces induce subtle, brace-specific and lateralized changes in plantar pressure distribution, potentially reflecting altered postural control strategies. Although limited to specific variables, these effects may be clinically relevant for fall risk assessment and individualized knee brace prescription in older adults. Full article

As diets shift towards more plant-based patterns, nutrients mainly supplied by animal-sourced foods are receiving greater attention. Among these are sulphur-containing amino acids (SCAAs) such as taurine, methionine, and cysteine. These compounds play important roles in neuroprotection, antioxidant defence, and cellular signalling; functions that are closely linked to cognitive health. This systematic review examined the effects of SCAA supplementation on cognitive performance in randomised controlled trials (RCTs). Eight RCTs involving 244 healthy participants met the inclusion criteria. All trials focused exclusively on taurine; no studies were found that tested methionine or cysteine. Each used an acute, single-dose design, assessing key cognitive domains and mood outcomes. Overall, acute doses of taurine (typically 1–3 g, up to ~50 mg/kg) produced, at best, small and inconsistent improvements in cognitive function. Most cognitive outcomes showed no effect. Trials that combined taurine with caffeine showed more reliable performance benefits, but they did not isolate taurine’s independent effects. Similarly, any positive effects on mood or well-being were minor, inconsistent, and typically observed only under specific conditions, such as when taurine was combined with caffeine, exercise, or sleep deprivation. Importantly, none of the studies measured participants’ habitual diets, baseline SCAA status, or specifically recruited individuals with low intake of animal-source foods. This means the cognitive effects of reduced SCAA intake in plant-based diets remain unknown. Current evidence from acute taurine trials provides limited support for short-term benefits to cognition or mood. Longer-term, well-designed studies are urgently needed. These should assess habitual diet and baseline SCAA status and focus on populations with lower animal-derived food intake. Only then can we determine whether lower SCAA availability in plant-based diets represents a nutritional ‘green gap’ with implications for brain health. Full article

The cement industry is a major source of anthropogenic CO₂ emissions due to its energy-intensive production process and calcination of limestone. Producing one ton of cement emits approximately one ton of CO₂, and cement accounts for about 5% to 8% of global CO₂ emissions. In this context, cement-less one-part (“just-add-water”) ambient-cured geopolymer concrete (GPC) has gained attention due to its environmental friendliness and practicality for large-scale cast-in-situ construction. However, field adoption remains limited, mainly due to the scarcity of data on mechanical properties and durability, as well as the lack of widely accepted standards and specifications. This paper is part of the larger research on tensile performance of anchors embedded in GPC. It is well understood that the tensile performance of anchors installed in concrete substrate is largely influenced by their effective embedment depth and the substrate’s mechanical properties, particularly the fracture energy and modulus of elasticity. Therefore, prior to the investigation of the tensile performance of anchors in GPC, it is crucial to understand the mechanical behaviour of the GPC substrate itself. This study examined key parameters that influence the compressive strength of one-part ambient-cured slag/fly ash-based GPC. The alkali content, slag content, water-to-solid (W/S) ratio, and aggregate content were investigated. Additionally, various mechanical properties such as uniaxial tensile strength, splitting tensile strength, elastic modulus, and fracture energy of the hardened GPC are presented. The test results revealed that higher slag and activator content enhanced compressive strength, whereas a higher aggregate content reduced the strength. The strength gain was also attributed to higher alkali content, lower W/S ratio, and increased binder content; however, excessive alkali and an overly low W/S ratio caused rapid setting due to accelerated reaction kinetics. The 7-day compressive strength ranged from 62% to 78% of the 28-day strength, while there was no notable strength gain after 28 days of curing. The developed GPC attained compressive strengths of over 40 MPa at 28 days and 50 MPa at 56 days. The uniaxial tensile strength test demonstrated a ratio of 0.87 relative to splitting tensile strength. The findings also indicated that the aggregate conditions and curing regimes (whether using as-is aggregates with moisture curing or oven-dried aggregates with sealed curing) had no meaningful effect on the mean compressive strength of GPC and its reproducibility. Full article

This study proposes a topic-modeling guided framework that enhances image classification by introducing semantic clustering prior to CNN training. Images are processed through two key-point extraction pipelines: Scale-Invariant Feature Transform (SIFT) with Sobel edge detection and Block Gabor Filtering (BGF), to obtain local feature descriptors. These descriptors are clustered using K-means to build a visual vocabulary. Bag of Words histograms then represent each image as a visual document. Latent Dirichlet Allocation is applied to uncover latent semantic topics, generating coherent image clusters. Cluster-specific CNN models, including AlexNet, GoogLeNet, and several ResNet variants, are trained under identical conditions to identify the most suitable architecture for each cluster. Two topic guided integration strategies, the Maximum Proportion Topic (MPT) and the Weight Proportion Topic (WPT), are then used to assign test images to the corresponding specialized model. Experimental results show that both the SIFT-based and BGF-based pipelines outperform non-clustered CNN models and a baseline method using Incremental PCA, K-means, Same-Cluster Prediction, and unweighted Ensemble Voting. The SIFT pipeline achieves the highest accuracy of 95.24% with the MPT strategy, while the BGF pipeline achieves 93.76% with the WPT strategy. These findings confirm that semantic structure introduced through topic modeling substantially improves CNN classification performance. Full article

Wrist-worn devices enable new paradigms of implicit and continuous user authentication; however, identifying biometric modalities that combine reliability with practical integrability remains challenging. Inner-wrist skin texture represents a relatively unexplored biometric characteristic that may be acquired unobtrusively using commodity hardware. This study evaluates biometric verification based on inner-wrist skin texture using an off-the-shelf capacitive fingerprint sensor and an unmodified, manufacturer-provided fingerprint verification algorithm. Two experiments were conducted. Experiment 1 assessed baseline verification performance under controlled acquisition conditions in a cohort of 33 participants (21 male, 12 female; mean age 30.0 ± 16.9 years, range 10–71 years), yielding 1768 genuine authentication trials. Experiment 2 examined the effect of wrist posture variation under controlled flexion in a separate cohort of 15 participants (11 male, 4 female; mean age 30.9 years, range 18–49 years), with 3900 authentication trials recorded. Across 86,897 impostor comparisons in Experiment 1, no false acceptances were observed, corresponding to a conservative upper bound on the false acceptance rate of 6.7 × 10⁻⁵ at the 99.7% confidence level, while the false rejection rate was approximately 2.93%. In Experiment 2, the overall false rejection rate increased to 3.52%, with no clear monotonic relationship between wrist angle and verification performance within the tested range. The results demonstrate that inner-wrist skin texture can be captured and matched using fingerprint-oriented sensing and matching technology under controlled conditions, providing an experimental baseline for this biometric modality. At the same time, the use of a closed matching algorithm and a sensor designed for fingerprints limits interpretability and generalization. These findings motivate further investigation using dedicated recognition methods, larger sensing areas, and extended evaluation protocols tailored specifically to wrist skin print biometrics. Full article

Background/Objectives: Manual preparation of semisolid formulations (creams, ointments, gels) is prone to variability in mixing energy and time, which may compromise uniform API distribution. This study aimed to evaluate an Electronic Mortar and Pestle (EMP; Unguator™) as a standardized compounding tool, with objectives to: (i) validate stability-indicating UHPLC methods; (ii) assess content uniformity across jar strata; (iii) quantify the impact of mixing time and rotation speed via design of experiments (DOE); and (iv) verify cleaning effectiveness and cross-contamination risk. Methods: Five representative formulations were compounded: urea 40%, clobetasol 0.05%, diclofenac 2.5% in hyaluronic acid 3% gel, urea 10% + salicylic acid 1%, and hydroquinone 5%. UHPLC methods were validated per ICH Q2(R2) and stress-tested under acid, base, oxidative, thermal, and UV conditions. Homogeneity was assessed by stratified sampling (top/middle/bottom). A 3² factorial DOE (time: 2/6/10 min; speed: 600/1500/2400 rpm) modeled effects on % label claim and RSD. Cleaning validation employed hydroquinone as a tracer, with swab sampling pre-/post-use and post-sanitization analyzed by HPLC. Results: All UHPLC methods met specificity, linearity, precision, accuracy, and sensitivity criteria and were stability-indicating (Rs ≥ 1.5). Formulations achieved 90–110% label claim with strata CV ≤ 5%. DOE revealed speed as the dominant factor for clobetasol, urea, and diclofenac, while time was more influential for salicylic acid; gels exhibited curvature, indicating diminishing returns at high rpm. Model-predicted optima were implementable on the Unguator™ with minor rounding of rpm/time. Cleaning validation confirmed post-sanitization residues below LOQ and <10 ppm acceptance. Conclusions: The Unguator™ provides a practical, parameter-controlled route for compounding pharmacies to standardize semisolid preparations, achieving reproducible layer-to-layer content uniformity within predefined criteria under the evaluated conditions through programmable set-points and validated cycles. DOE-derived rpm–time relationships define an operational design space within the studied ranges and support selection of implementable device settings and set-points. Importantly, the DOE-derived “optima” in this study are optimized for assay-based content uniformity (mean % label claim and strata variability). Cleaning validation supports a closed, low-cross-contamination workflow, facilitating consistent routines for both routine and complex formulations. Overall, the work implements selected QbD elements (QTPP—Quality Target Product Profile; CQA—Critical Quality Attribute definition; CPP—Critical Process Parameter identification; operational design space; and a proposed control strategy) and should be viewed as a step toward broader lifecycle QbD implementation in compounding. Full article

Background/Objectives: Diabetes is major metabolic disorder and rapidly increasing public health problem globally. The greatest way to reduce diabetic complications is adequate knowledge about the condition. Hence, the primary objectives of this study were to evaluate the psychometric properties of the Simplified Diabetes Knowledge Test—Arabic version (SDKT-A) among Iraqi insulin-dependent diabetic patients. Additionally, the secondary objectives were to assess the associated independent variables and the risk of atherosclerosis and cardiovascular risk event by using atherogenic indices and lipid ratios with the SDKT-A. Methods: A cross-sectional, descriptive study was conducted in primary healthcare clinics. The SDKT was translated into Arabic using forward–backward translation, reconciliation, and pilot testing. Thereafter, psychometric properties of the SDKT-A were evaluated depending on different criteria. Atherogenic indices of Castelli risk indices I and II (CRI-I and II), triglyceride/HDL ratio, non-HDL-C ratio, atherogenic coefficient (AC), and triglyceride–total cholesterol–body weight index (TCBI) were calculated using specific formulas. Results: The SDKT-A questionnaire showed acceptable readability and validity. Cronbach’s alpha test (95% confidence interval) was 0.662 (0.59–0.73). The Pearson correlation coefficient of reliability for test–retest was found to be 0.659. The item difficulty index for most items was between 0.237 and 0.877. The point biserial correlation values ranged from 0.028 to 0.535 with Ferguson’s sigma value equal to 0.962. The content validation results showed a significant content validity ratio (CVR) value for most of the questions, ranging from 0.8 to 1. The content validity index (CVI) value for SDKT-A was found to be 0.98, which showed good agreement between experts. In addition, the exploratory factor analysis with promax rotation identified four domains for the final 20 items of the SDKT-A that explained 41.83% of the scale total variance. The mean score of the SDKT-A was 11.09 ± 3.40. The total score of the SDKT-A was positively and significantly correlated with education level (r = 0.322, p < 0.01). In addition, the total scores of the SDKT-A were negatively and significantly correlated with glycemic control, age, CRI-I, CRI-II, triglyceride/HDL ratio, AC, non-HDL-C ratio, and TCBI. Furthermore, the glycemic control (HbA1c) was positively and significantly correlated with the preventive measures factor (r = 0.175, p < 0.05), and were negatively and significantly correlated with the lifestyle and modification factor (r = −0.169, p < 0.05), diet and monitoring factor (r = −0.158, p < 0.05), and awareness factor (r = −0.149, p < 0.05). Conclusions: This study showed acceptable psychometric properties for the SDKT-A, with low levels of knowledge of diabetic disease in the sample population. Finally, comprehensive and interactive educational programs regarding lifestyle and modification, diet, and monitoring and awareness in primary healthcare centers in Iraq are warranted. Full article

This study applies phase plane analysis of medio-lateral, anteroposterior, and vertical directions for the coordination assessment of whole-body (WB) center of mass (COM) movement during the impulse phase of a standard maximum vertical jump (MVJ) with long, short, and no countermovement (CM). A video system and force platform were used, with the amplitudes of WB COM excursion obtained from image-based motion capture at each anatomical direction, and the 2D and 3D mean radial distance were compared under long, short, and no CM conditions. The estimate of the population mean length was used as a measure of distribution concentration, and the Rayleigh statistical test for circular data was applied with the sample distribution critical value. Watson’s U² goodness-of-fit test for the von Mises distribution was used with the mean direction and concentration factor. The applied metrics led to the detection of shared and specific features in the global and phase plane analysis of WB COM movement coordination in the medio-lateral, anteroposterior, and vertical directions during long, short, and no CM conditions in relation to MVJ performance assessed from ground reaction force (GRF) through the force platform. Thus, long, short, and no CM impulses share lower amplitudes of WB COM excursion in the medio-lateral direction and mean radial distance to its mean, whereas the anteroposterior and vertical excursion of WB COM, along with the 2D transversal and 3D spatial length of the WB COM path, present as potential predictors of MVJ performance, with distinct behavior in long CM compared to short and no CM. Additionally, the applied workflow on generalized phase plane analysis led to the detection, through complementary metrics, of the anatomical WB COM movement directions with higher coordination based on phase concentration tests at 5% significance, in line with MVJ performance under different CM conditions. Full article

Reliable state estimation is critical for the safe operation of second-life battery systems but is severely hindered by significant parameter heterogeneity arising from diverse historical aging conditions. Traditional static models struggle to adapt to such variability, while online identification methods are prone to divergence under dynamic loads. To overcome these challenges, this paper proposes a Dual-Anchor Adaptive Fusion Framework for robust State of Charge (SOC) estimation and State of Health (SOH) soft-sensing. Specifically, to establish a reliable physical baseline, an automated Dynamic Relaxation Interval Selection (DRIS) strategy is introduced. By minimizing the fitting Root Mean Square Error (RMSE), DRIS systematically extracts high-fidelity parameters to construct two “anchor models” that rigorously define the boundaries of the aging space. Subsequently, a residual-driven Bayesian fusion mechanism is developed to seamlessly interpolate between these anchors based on real-time voltage feedback, enabling the model to adapt to uncalibrated target batteries. Concurrently, a novel “SOH Soft-Sensing” capability is unlocked by interpreting the adaptive fusion weights as real-time health indicators. Experimental results demonstrate that the proposed framework achieves robust SOC estimation with an RMSE of 0.42%, significantly outperforming the standard Adaptive Extended Kalman Filter (A-EKF, RMSE 1.53%), which exhibits parameter drift under dynamic loading. Moreover, the a posteriori voltage tracking residual is compressed to ~0.085 mV, effectively approaching the hardware’s ADC quantization limit. Furthermore, SOH is inferred with a relative error of 0.84% without additional capacity tests. This work establishes a robust methodological foundation for calibration-free state estimation in heterogeneous retired battery packs. Full article

Background/Objectives: Accurate hydration assessment is critical for optimizing performance and preventing heat-related complications in ultra-endurance athletes. This study evaluated the agreement and reliability between urine reagent strips and refractometry for field-based hydration assessment via urine-specific gravity (USG) in ultra-trail runners. Methods: Thirty-four ultra-trail runners (22 males, 12 females; mean age 43.71 ± 11.50 years) participated during The Coastal Challenge, a 241-km multi-stage ultra-trail competition. Urine samples were collected before and after the first two stages (Stage 1: 41 km, 1071 m elevation; Stage 2: 40 km, 1828 m elevation). USG was measured using semi-quantitative urine reagent strips (Combur10Test M) and a handheld digital refractometer (Palm Abbe™). Agreement was assessed via paired t-tests, Pearson and Spearman correlations, intraclass correlation coefficients, and Bland-Altman plots across four measurement time points. Results: Strong agreement existed between methods with correlation coefficients of 0.92–0.99 (p < 0.01) within the hydration range typical of well-prepared ultra-endurance athletes (USG 1.010–1.020). No significant differences were found between devices at any time point (all p > 0.05). Bland-Altman analyses revealed minimal mean bias (range: −0.002 to +0.001 g/mL) and narrow limits of agreement, with fewer than 5% of values falling outside limits. Both methods detected significant increases in USG from pre- to post-stage (p < 0.01), indicating exercise-induced hypohydration. Conclusions: Semi-quantitative urine reagent strips and handheld refractometers demonstrate strong agreement for hydration assessment in ultra-trail runners under field conditions when not severely hypohydrated, supporting their interchangeable use for practical monitoring. Full article

This study proposes and evaluates a data-driven framework for short-term System Marginal Price (SMP) forecasting in the Chilean National Electric System (NES), a power system characterized by high penetration of variable renewable generation and persistent transmission congestion. Using publicly available hourly operational data for 2024, multiple machine learning regressors including Linear Regression (base case), Bayesian Ridge, Automatic Relevance Determination, Decision Trees, Random Forests, and Support Vector Regression are implemented under a node-specific modeling strategy. Two alternative approaches for predictor selection are compared: a system-wide methodology that exploits lagged SMP information from all network nodes; and a spatially filtered methodology that restricts SMP inputs to correlated subsystems identified through nodal correlation analysis. Model robustness is explicitly assessed by reserving January and July as out-of-sample test periods, capturing contrasting summer and winter operating conditions. Forecasting performance is analyzed for representative nodes located in the northern, central, and southern zones of the NES, which exhibit markedly different congestion levels and generation mixes. Results indicate that non-linear and ensemble models, particularly Random Forest and Support Vector Regression, provide the most accurate forecasts in well-connected areas, achieving mean absolute errors close to 10 USD/MWh. In contrast, forecast errors increase substantially in highly congested southern zones, reflecting the structural influence of transmission constraints on price formation. While average performance differences between M1 and M2 are modest, a paired Wilcoxon signed-rank test reveals statistically significant improvements with M2 in highly congested zones, where M2 yields lower absolute errors for most models, despite relying on fewer inputs. These findings highlight the importance of congestion-aware feature selection for reliable price forecasting in renewable-intensive systems. Full article