Search Results (1,397)

Press-pack insulated-gate bipolar transistor (IGBT) devices are key components of modular multilevel converter (MMC)-based HVDC systems, yet the parallel connection of tens of chips inside one sealed housing makes their internal aging difficult to monitor—existing methods cannot determine which chip or thermal interface has degraded without invasive sensing. This paper proposes a two-layer aging condition monitoring framework based solely on non-invasive external case temperature measurements, comprising an online detection layer and a detection layer. For the diagnosis layer, a coupled thermal network state-space model of the press-pack module is established; its observability matrix is shown to be of full rank, proving that aging at any internal location is detectable from external case temperatures, and a steady-state sensitivity analysis reduces the required measurement to the module-center case-temperature pair and yields a failure-signature rule that discriminates five aging modes. A 3D electrothermal finite-element model, validated against fiber Bragg grating (FBG) measurements on an MMC sub-module with deviations below 1.2 °C at all load levels, is then used to build an aging fingerprint database of 270 scenarios that maps a measured case-temperature distribution to the aging location, the affected component, and the severity of any cooling-system degradation. For the detection layer, a deep neural network (DNN) trained only on healthy-state data provides online early warnings through its prediction residual. Experiments on a 2.2 kV/2100 A MMC sub-module platform and a solid-state DC–DC transformer platform confirm that cooling-system degradation is reliably detected and that the detection layer transfers across converter types. Even in the least sensitive aging case, the predicted case-temperature signature (0.036 °C) exceeds the 0.01 °C resolution of standard thermocouple instrumentation, which indicates that genuine device aging would be resolvable as well. The framework provides a non-invasive online health-monitoring solution that can also locate the degraded element in high-power press-pack devices. Full article

Using daily rainfall observations from 1961 to 2022 (62 years) across 15 typical sub-regions within the water-receiving area of the Zhejiang East Water Diversion Project, this study systematically examines the spatiotemporal distribution, long-term trends, multi-scale variability, and inter-regional correlations of rainfall. An integrated framework was employed, including the Mann–Kendall (MK) trend test, Sen’s slope estimator, Hurst exponent analysis, and multi-scale sliding window analysis. Spatially, the multi-year average daily rainfall ranges from 3.47 mm to 5.68 mm, following a distinct “high in the center, low in the west” pattern, with the Yuyao Plain Mazhu Midstream Area identified as the regional rainfall maximum. In the raw Mann–Kendall test all 15 sub-regions exhibit increasing trends, of which 14 remain statistically significant (p < 0.05) after trend-free pre-whitening; Sen’s slopes follow a “coastal > riverine > hilly” gradient. Hurst exponents greater than 0.5 suggest persistent behavior in the rainfall series, although attribution to external forcing or separation from low-frequency climate variability requires additional analysis. Multi-scale sliding window analysis reveals strong scale dependence: the amplitude of trend fluctuations decreases by approximately 90% from ±16 mm at the 3-month scale to ±1.7 mm at the 12-month scale, while inter-regional correlation coefficients increase from 0.83 to 0.87. Notably, the Yuyao Plain Mazhu Midstream Area displays a unique “increase-then-decrease” correlation pattern, reflecting its distinctive hydro-geographic conditions. These findings provide a scientific basis for climate-adaptive scheduling of water diversion projects, supporting water security, urban resilience, and climate action. Full article

Background: Soft tissue sarcomas (STS) in children, adolescents, and young adults are rare and biologically heterogeneous, and prognostic factors for local recurrence, metastasis, and survival remain incompletely defined. This study aimed to characterize oncological outcomes and to explore clinical, surgical, and treatment-related factors associated with these endpoints at a tertiary sarcoma center. Methods: This retrospective single-center cohort study included 42 consecutive patients aged ≤25 years with histologically confirmed STS who underwent definitive surgery between June 2018 and November 2025. Local recurrence-free (LRFS), metastasis-free (MFS), event-free (EFS), and overall survival (OS) were estimated by the Kaplan–Meier method with competing-risks sensitivity analyses. Given the limited number of events, pre-specified univariable comparisons (log-rank, Fisher exact, Mann–Whitney U) were performed and regarded as exploratory. Results: The median age was 16.7 years and 26 patients (61.9%) were female. A microscopically complete (R0) resection was achieved in 35 of 41 patients with assessable margins (85.4%), and limb preservation in 38 (90.5%). Over a median follow-up of 23.1 months, the 24- and 60-month estimates were LRFS 85.7% and 79.6%, MFS 93.3% and 84.8%, EFS 82.3% and 64.7%, and OS 96.0% and 81.5%, respectively. Metastatic disease at diagnosis was the only variable associated with metastatic status at last follow-up (p = 0.035). No factor was robustly associated with local recurrence; anatomical site (p = 0.68), resection-margin status (p = 0.23), and tumor volume (p = 0.97) were not significant, and an apparent association with a sub-millimeter margin rested on a single event (complete separation). Conclusions: Metastatic disease at diagnosis emerged as the most important prognostic factor for MFS in our cohort. Despite the complexity of treatment in this young patient population, high rates of complete resection and limb preservation were achieved, highlighting the value of multidisciplinary management in specialized sarcoma centers. While the limited cohort size did not allow the identification of further statistically significant prognostic factors for local recurrence, such associations may become detectable in larger multicenter studies. Full article

The rapid expansion of artificial intelligence (AI) has significantly increased the power demand of high-performance data centers. This study constructs an integrated energy system based on Small Modular Reactors (SMRs). The system adopts a supercritical carbon dioxide (S-CO₂) Brayton cycle that directly supplies power for data centers. This work outlines the deep coupling of three core modules: power generation, S-CO₂ energy storage, and waste heat absorption refrigeration driven by residual heat. The design enables coordinated optimization and cascaded utilization of nuclear multi-energy flows. We first build a full thermodynamic model for the whole system, and then formulate dynamic operation scheduling strategies. A 24 h full-condition simulation is carried out. The simulation object is a 125 MW SMR with a 100 MW data center in an off-grid island operation mode. Simulation results for the key performance indicators are as follows: The system cycle thermoelectric conversion efficiency reaches 45.00%. Compared with equal-capacity SMR units equipped with traditional steam cycles, the efficiency rises by 12.5 percentage points. The overall comprehensive energy efficiency hits 82.21%, and the system exergy efficiency is 74.30%. The system achieves a completely self-sufficient power supply without grid support. Its load power deficit rate is only 1.73%. Two operation modes dominate daily system operation: surplus power charging for peak shaving accounts for 60.6% of total runtime, while energy discharging to fill power deficits equates to 38.4%. Waste heat refrigeration requires roughly 93% of the data center’s total cooling demand. This research provides a new technical framework for low-carbon and sustainable construction of next-generation high-performance data centers. The integrated system we propose provides a replicable zero-carbon off-grid energy technical route. It can serve large computing hubs constructed under China’s national “East Data, West Computing” strategy. Full article

Background: Retrograde intrarenal surgery (RIRS) is a well-established minimally invasive technique for the management of renal stones. However, intraoperative physiological and anesthetic events during the procedure remain poorly characterized, particularly in the context of newer technologies such as suction-assisted systems. Objective: The objective of this study is to evaluate the occurrence of intraoperative anesthetic and physiological adverse events during suction-assisted RIRS and to assess its safety profile. Methods: We conducted a prospective, single-center observational sub-analysis of the “Trifecta factors in ureteroscopy” project. A consecutive cohort of 30 patients undergoing RIRS with suction ureteral access sheaths (sUAS) between August 2025 and January 2026 was analyzed. Baseline demographics, stone characteristics, and intraoperative anesthetic parameters—including airway pressure, Bispectral Index (BIS), and hemodynamic events—were prospectively recorded. Primary endpoints included intraoperative fever, sustained tachycardia, blood pressure alterations, and anesthetic complications. Results: The cohort had a mean age of 56.4 years with an equal gender distribution. Mean stone size was 12.9 mm, and mean operative time was 50.3 min. Intraoperative monitoring demonstrated stable anesthetic conditions, with a mean airway pressure of 16.3 cmH₂O and a mean BIS of 42.6. No major anesthetic complications or significant blood pressure fluctuations were observed. Minor intraoperative events included fever in 10% of cases and sustained tachycardia in 6.7%. No cases of intraoperative anesthetic instability were recorded. Conclusions: Within this descriptive cohort, suction-assisted RIRS under general anesthesia demonstrated a favorable safety profile with stable intraoperative parameters, though larger comparative studies are required to confirm these preliminary findings. Full article

Pepper breeding in China faces challenges including narrow genetic backgrounds, variety homogenization, and authenticity issues in the seed market. The original SSR-based variety identification system (NY/T 2475-2013) has limitations such as low polymorphism and uneven marker distribution, failing to meet current efficiency requirements. In this study, 240 pepper varieties were used. Fluorescent capillary electrophoresis was employed to analyze issues in the original system. A total of 150 SSR primer pairs was screened using eight phenotypically distinct varieties, followed by rescreening with 96 varieties from different origins. After marker optimization, a new identification system was established, validated using 144 varieties, and a DNA fingerprint database was constructed. Phenotypic experiments were conducted to evaluate applicability. The new system comprises 26 core primer pairs: 12 newly selected and 14 retained from the original standard. These primers are evenly distributed across chromosomes, with annealing temperatures optimized to 57 °C and grouped into seven multiplex panels. Across 240 pepper varieties, the 26 primer pairs detected 153 allelic variants and 293 genotypes. Average polymorphism information content (PIC) increased from 0.43 to 0.49 compared to the original system, and the discrimination efficiency reached 99.975% among 28,680 pairwise variety combinations. Phylogenetic and population structure analyses divided the 240 pepper varieties into six groups. Varieties with molecular genetic distance approaching zero showed high phenotypic similarity. The newly developed SSR-based system demonstrated excellent performance and strong applicability. It can be used for authenticity and purity identification, parentage testing, and auxiliary selection in DUS testing, supporting new variety rights protection and providing technical support for standardized management of the pepper seed industry. Full article

Curbing industrial air pollution remains central to China’s sustainable transition under the dual-carbon commitment, and the digital economy has been proposed as an enabler of cleaner production. Yet most evidence centers on carbon emissions and single-dimensional digital measures, leaving the link between multi-dimensional digital development and sulfur dioxide (SO₂) emissions underexplored. We construct a multi-dimensional digital economy index (DEI) from three county-aggregated digital integration sub-indicators using principal component analysis and assemble a balanced panel of thirty provinces over 2000–2022 (N = 690). Using two-way fixed-effects estimation with cluster-robust inference, we find that DEI is negatively associated with provincial SO₂ emissions: a one-unit increase in DEI is associated with 19.2 percent lower emissions. The association survives wild cluster bootstrap inference, province-specific linear trends, alternative composite weights, two alternative SO₂ measures, and a Bartik-style exposure–shock instrumental-variable sensitivity analysis. Integration-type dimensions carry the stronger effect. A formal mediation analysis shows that digital development lowers coal reliance and raises regulation and innovation. Yet no observable channel statistically mediates the total effect. This pattern is consistent with abatement operating through emissions monitoring and compliance. The findings position digital development as a correlate of air-quality improvement and inform digital infrastructure planning under the dual-carbon framework. Full article

Little is known regarding in-hospital transthoracic echocardiography (TTE) utilization and its prognostic implications among stable patients with acute myocardial infarction (AMI). We aimed to explore patient and disease characteristics, treatment strategies, and mid-term outcome following uncomplicated AMI according to TTE use during the hospitalization phase. A single-center, retrospective analysis was conducted that included consecutive adult individuals admitted for AMI who did not develop cardiogenic shock and who survived the index hospitalization. Stratified by in-hospital TTE administration status, the cohort was evaluated for all-cause mortality at 1-year post-discharge. Overall, 15,971 subjects (mean age 66 ± 14 years, 69.8% males, 46.1% with ST-elevation myocardial infarction) were analyzed, of whom 12,610 (79.0%) underwent TTE. TTE use correlated with younger age, fewer comorbidities, greater odds of invasive revascularization and intensive coronary care unit management, and lengthier hospital stay. Ultimately, it was associated with a lower rate, cumulative incidence, and—independent of accompanying prognostic markers—risk of all-cause mortality (n = 1032/12,619, 8.2% vs. n = 804/3361, 23.9%, p < 0.001; Log-Rank p < 0.001; adjusted hazard ratio 0.75, 95% confidence interval 0.67–0.83, p < 0.001). Similar results were observed within a 6270-patient, 1-1 propensity score-matched sub-cohort. To conclude, in our experience, in-hospital TTE administered for stable AMI patients was associated with improved mid-term survival. Further research is needed to re-evaluate the present-day recommendation’s Level of Evidence C for its routine use. Full article

Hydrometeorological changes profoundly influence runoff generation and evolution in river basins. It is of great significance to carry out runoff prediction research to ensure water resources security and improve disaster prevention and mitigation capabilities. In this paper, the Ganhe River Basin in Northeast China was taken as the research object. Based on the hydrometeorological and runoff data from 1980 to 2022, a variety of statistical methods were used to systematically study the climate change, runoff evolution characteristics and driving mechanism of the basin. Combined with BP neural network model and CMIP6 climate scenario data, the future runoff changes were predicted. The results showed that the precipitation and relative humidity showed a downward trend, while the temperature, sunshine and evapotranspiration showed an upward trend during the study period. The runoff showed a non-significant upward trend, and an abrupt change occurred in 2009. After the abrupt change, the runoff increased by 38.7% compared with the baseline period. The change in land use was the most significant from 1990 to 2000, and the area of cultivated land increased significantly. Correlation analysis showed that precipitation was the dominant meteorological factor affecting runoff change, and the contribution rate of human activities was 88.51%, which was much higher than that of climate change. The BP neural network model demonstrated satisfactory simulation performance, and the training set and test set R2 reached 0.88 and 0.82, respectively. In the future, both temperature and precipitation will increase under different SSP scenarios. On this basis, the BP neural network prediction results show that the runoff of the basin is generally increasing, and the increase is the most significant under the high emission scenario, and the risk of extreme hydrological events may be further aggravated. These findings provide scientific support for water resources management and ecological conservation in the Ganhe River Basin. Full article

Against the backdrop of global climate change and carbon neutrality strategies, land use carbon emissions have become a prominent topic amid regional efforts toward low-carbon transformation. However, existing studies on land-use carbon emissions have predominantly focused on humid and economically developed regions, while the unique carbon metabolism pathways of arid oasis–desert ecosystems, which are characterized by extremely low environmental carrying capacity and high sensitivity to land-use disturbance, remain largely unexplored. This study takes the oasis urban cluster in the Tarim Basin in southern Xinjiang Uygur Autonomous Region as the research object. This region belongs to a typical oasis–desert composite ecosystem, with a simple structure and low environmental carrying capacity (reflected by sparse vegetation cover < 20%, annual precipitation < 100 mm, extremely limited water resources, and high sensitivity to land disturbance). Its carbon metabolism pathway (i.e., the dynamic balance between carbon sources and sinks induced by land-use change) is fundamentally different from that in humid areas, and thus merits dedicated investigation. This study selects the period from 2000 to 2020 as the research period, which completely covers the acceleration period of urbanization and agricultural expansion in the Tarim Basin oasis urban cluster since the advancement of China’s Western Development Initiative. The data have a temporal resolution of 5 years (samples in 2000, 2005, 2010, 2015, 2020) and a spatial resolution of 30 m for land use and prefecture level for socio-economic indicators. Based on this, to fill the above-mentioned research gap, a research framework integrating the carbon emission coefficient accounting method, landscape pattern index, spatial autocorrelation analysis and geographic detector is adopted. Specifically, this study aims to systematically quantify the spatio-temporal evolution of land use carbon emissions and identify the most robust driving factors in the Tarim Basin oasis urban cluster by integrating multiple models, an approach that has not been previously applied to arid oasis regions. The research results show: (1) Based on the carbon emission coefficient method, total carbon emissions increased from 1.4455 million tons to 22.364 million tons, following a ‘slow-then-fast’ trajectory. In terms of temporal evolution, the study period can be further divided into three sub-stages: 2000–2005 (slow diffusion, with emission center skewed toward the northern energy-intensive zone), 2005–2015 (rapid restructuring, characterized by a ‘unipolar surge’ in Aksu and spread to the central oasis belt), and 2015–2020 (high-intensity stabilization, forming a cross-regional emission belt). Meanwhile, the land use structure has undergone a significant transformation. Construction land and cultivated land have continued to expand, while ecological land has significantly shrunk, resulting in a complex transformation pattern of oasis–desert ecotone. (2) The overall landscape became increasingly fragmented and diversified, the integrity of ecological space was damaged, and the regional carbon sink function was weakened. (3) The spatial autocorrelation analysis indicates that the spatial distribution of carbon emissions shows a heterogeneous pattern, forming a high-emission concentration area centered around Aksu-Bayingol. However, the global Moran’s I index is negative (such as −0.171 in 2020, p > 0.05), suggesting that carbon emissions have not formed a significant spatial clustering. (4) Carbon emissions are dominated by human and economic factors, and the interaction of factors is significant. The geographic detector identifies population density (average q value 0.904) and the proportion of construction land (average q value 0.858) as the key determinants of spatial variation in carbon emissions, reflecting the sensitive response of the human-nature system of arid zones to the urbanization process. These findings not only clarify the spatio-temporal features and driving forces of land use carbon emissions in the Tarim Basin oasis urban cluster, but also provide a replicable analytical framework for carbon-emission research in other arid and semi-arid regions worldwide. Based on these findings, we discuss the unique driving mechanisms of carbon emissions in arid regions, conclude that construction land expansion and population density are the dominant factors, and recommend a three-tier zoning governance system (carbon source control zone, carbon sink enhancement zone, coordinated development zone) for low-carbon spatial planning in arid areas. Full article

Objective: To investigate the occurrence trends, distribution characteristics, and contributing factors of medical damage liability disputes in Chongqing from 2012 to 2024, thereby providing data support for risk prevention and control in regional medical institutions and for health policy formulation. Methods: A retrospective study design was employed. Adjudication documents pertaining to medical damage liability disputes in Chongqing, spanning from January 2012 to December 2024, were retrieved from the China Judgments Online database. A total of 1761 eligible cases were selected based on standardized criteria. Statistical analysis was conducted on core case variables. Results: The number of cases exhibited a fluctuating upward trend from 2012 to 2024, peaking in 2019 (276 cases) before subsequently declining. The average actual compensation amount demonstrated a steady increase, reaching its highest level in 2023 (259,200 ± 53,900 CNY). Deficiencies in diagnostic and therapeutic procedures constituted the predominant cause of disputes (50.37%). Surgical departments (46.05%) and orthopedics departments (22.60%) were identified as high-risk specialties. Tertiary public hospitals accounted for the highest proportion of disputes (48.89%) and the highest average actual compensation (216,100 ± 10,600 CNY). Regarding the apportionment of liability, secondary liability attributed to the medical provider was the most frequent (29.59%). Patient death was the most common adverse outcome (37.14%). The highest average compensation amount was associated with persistent vegetative state outcomes (777,500 ± 157,200 CNY). Statistically significant differences in actual compensation amounts were observed across all outcome categories (all p < 0.05). Conclusions: The situation concerning medical damage disputes in Chongqing remains severe. The etiologies of disputes are predominantly attributable to medical technical factors—particularly deficiencies in diagnostic and therapeutic procedures—and failure to fulfill a duty of care, both of which are associated with elevated compensation amounts. Given that this study employed a retrospective descriptive design, the aforementioned findings reflect associations among variables rather than causal relationships. Based on the identified risk hierarchy, it is recommended that priority be given to strengthening perioperative safety verification and re-certification of critical procedural skills in high-incidence departments, such as surgery and obstetrics and gynecology, enhancing informed consent and medical record quality management, and advancing the tiered healthcare delivery system, with a view to reducing medical risks and mitigating doctor–patient conflicts. Full article

Pyropia haitanensis polysaccharides have attracted growing attention for their diverse biological activities. In this study, we developed a synergistic extraction approach combining ultrasonic-assisted treatment and enzymatic hydrolysis using cellulase and pectinase. Response surface methodology (RSM) was applied to optimize the extraction conditions, which were determined as follows: 1.48% cellulase, 1.47% pectinase, 180 W ultrasonic power, and 65.9 °C temperature. Under these conditions, the polysaccharide yield reached 10.184 ± 0.27%. The crude extract was then purified through sequential DEAE Sepharose FastFlow and Sephadex G-75 chromatography, resulting in the purified fraction PPHP3. Monosaccharide analysis revealed that galactose, glucose, and glucuronic acid constituted the primary components in a molar ratio of 98.3:0.46:1.24. This polysaccharide exhibited a weight-average molecular weight of 25.208 kDa, a sulfate content of 8.64 ± 0.05%. In hypolipidemic assays using oleic acid-induced HepG2 cells, PPHP3 significantly reduced intracellular triglycerides (TG), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL-C), while simultaneously increasing HDL-C levels. These findings highlight the potential of P. haitanensis polysaccharides for hypolipidemic applications and establish a scientific foundation for their development in therapeutic and practical contexts. Full article

Three-dimensional surface reconstruction is essential for accurately acquiring the external quality parameters of watermelons, such as size, volume, and defect area. Binocular stereo vision provides a low-cost and easily deployable solution for the single-view 3D reconstruction of watermelons. However, watermelons present highly similar surface textures, and as typical spheroid-like objects, the excessive angle between surface normals of edge regions and the camera optical axis leads to insufficient feature representation. Consequently, directly applying existing stereo matching algorithms often introduces matching ambiguities, and lightweight networks struggle to balance real-time performance with matching accuracy. This study focuses on the high-precision single-view point cloud generation of Kirin watermelons. To address these issues, we first construct a cross-modal, high-precision Kirin watermelon stereo matching dataset. Building upon the Fast-ACVNet+ architecture, we then propose MI-ACVNet, a lightweight stereo matching network tailored for high-precision watermelon point cloud acquisition. In the feature extraction stage, a Multi-Scale Stereo Feature Extraction (MSFE) module is adapted. By incorporating the re-parameterized network MobileOne and Epipolar-Enhanced Coordinate Attention (E2CA), MSFE improves the discriminative capability for weak and similar textures without compromising inference speed. For cost computation, a Coarse-to-Fine Cascaded Residual Correction (C2F-CRC) strategy is incorporated to construct a fine-grained cost volume via sub-pixel interpolation, enhancing the network’s ability to capture subtle surface fluctuations. Furthermore, a Semantics-Guided Region-Aware Loss (SGRA-Loss) is formulated, leveraging semantic masks to apply differentiated supervision weights across edge, center, and background regions to significantly improve edge matching accuracy. Ablation studies validate the effectiveness of the MSFE, C2F-CRC, and SGRA-Loss components. Compared to the baseline model, the full MI-ACVNet reduces the End-Point Error (EPE) by 19.5% and the Bad-0.5 error rate by 34.5% in the watermelon region. Furthermore, when compared against five mainstream algorithms (StereoNet, AANet, HSMNet, LightStereo-L, and NMRF-swint), MI-ACVNet achieves state-of-the-art performance: EPE and Bad-0.5 are reduced to 0.091 pixels and 1.159%, respectively, with a single-frame inference time of only 46 ms. The average depth error of the reconstructed point clouds is merely 0.26 mm. By ensuring both real-time efficiency and high-precision depth estimation, this method demonstrates promising potential for deployment in industrial Kirin watermelon sorting lines, driving sorting equipment toward higher precision and intelligence. Full article

Tuning the d-band electronic structure of non-noble-metal catalysts is a central strategy for an alkaline hydrogen evolution reaction (HER), yet how composition controls the d orbital in multi-Wyckoff-site layered systems remains insufficiently understood. Here, layered (Fe_1-xNi_x)₃GaTe₂ single crystals (x = 0.2–1.0) were synthesized by the self-flux method as a platform to address this question. Single-crystal XRD and EDS confirm that Ni is uniformly incorporated into the parent P6₃/mmc framework while inducing a composition-dependent lattice evolution. Electrochemical measurements in 1.0 M KOH reveal a clear volcano-shaped composition dependence, peaking at x = 0.6, where the lowest overpotential, the smallest Tafel slope (94 mV dec⁻¹), the lowest charge-transfer resistance and the largest double-layer capacitance are simultaneously reached. First-principles calculations show that Ni doping reshapes the Fe-site d orbital strongly composition-dependent rate: the Fe d-band center upshifts rapidly by ~0.5 eV between x = 0.4 and x = 0.6, while the Ni d-band center stays nearly fixed in the same composition range. The maximum of HER activity therefore aligns with a steep upshift of the Fe d-band center rather than with the Ni content itself. Charge-density mapping of (Fe_0.4Ni_0.6)₃GaTe₂ further demonstrates that the electron-enriched regions are located on the Fe and interlayer Ni3 sublattices that dominate the d states near E_F. Full article

To overcome the limitations of single-dimensional data and low efficiency in traditional cycloidal gear inspection, a comprehensive machine vision-based method was proposed. A high-precision vision platform was established, and a Sigmoid surface-based edge detection algorithm was employed for sub-pixel edge localization. Logarithmic transformation combined with light intensity compensation was applied to correct saturation-induced errors. The pixel equivalent and compensation coefficient were systematically calibrated using a dot-matrix plate and gauge blocks. A sub-pixel tooth profile model in the physical coordinate system was reconstructed through pixel equivalent calibration, dynamic light intensity compensation, and multi-coordinate transformation. Comparative tests against a coordinate measuring machine (CMM) verified that the point-to-point deviation between the two measurement systems was within 10 μm (maximum 11.62 μm). The inherent tooth profile deviation of the tested cycloidal gears, which reflects the machining quality of workpieces, ranged from 24 μm to 37 μm. Multiple repeated tests prove that the system achieves a repeat positioning accuracy of 0.8 μm. Based on the measurement characteristics, a hybrid analytical method integrating Cartesian and polar coordinate systems was developed, enabling the simultaneous evaluation of critical geometric tolerances, such as the diameters of the center hole and crankshaft hole. The full inspection cycle for cycloidal gears was reduced to 13 s, which demonstrates a substantial efficiency improvement over traditional methods. Full article