Inverse Analysis Research Articles (Page 1)

Water inrush disasters pose a serious threat during tunnel construction. Accurately evaluating their evolution process is essential for timely prevention and risk mitigation. Given the staged nature of seepage-instability-induced inrushes and the sensitivity of borehole resistivity imaging to water-bearing anomalies, this study explores the use of borehole resistivity methods to monitor the evolution of such events. A four-stage geoelectrical evolution model is developed based on the characteristics of inclined fault-related water inrushes. A time-lapse evaluation method combining least squares inversion and resistivity ratio analysis is proposed to assess the inrush process. Numerical simulations show that this method achieves a localization error below 2 m for inclined water-conducting channels. Across the four stages, the resistivity ratio of the channel ranges from 0.65 to 1.40, capturing the three-dimensional expansion of the inrush pathway. These findings confirm that borehole resistivity imaging effectively characterizes the evolution of water inrush disasters and supports early warning and mitigation strategies.

Association of Admission Nutritional Status Evaluated by the Controlling Nutritional Status Score with Periprosthetic Joint Infection after Primary Total Joint Arthroplasty: An Exploratory Analysis.

Abstract The Numanotaira Crater (NC) of the Adatara Volcano is located in northern Fukushima Prefecture, northeastern Japan, and has repeatedly undergone phreatic eruptions and volcanic lahars. Recent eruptions occurred between 1899 and 1900, and hydrothermal activity has continued since then. Based on previous geophysical and geological studies, a well-developed hydrothermal system is believed to have formed beneath the crater. However, the subsurface structure of shallow hydrothermal systems has not been investigated. In this study, an audio-frequency magnetotelluric survey was conducted in and around NC to clarify the distribution of shallow hydrothermal systems and understand the relationship between the subsurface structure and recent hydrothermal activities to examine the potential for phreatic eruptions at NC. In addition, several surveys (diffuse CO 2 flux, ground temperature, and resistivity of hot spring and river water) were conducted over a wide area of NC to assist in interpreting the resistivity structure. As a result of three-dimensional resistivity inversion analysis considering the steep topography around NC, the following findings were obtained. A low-resistivity zone was found beneath the crater floor, which had previously been the center of hydrothermal activity. However, part of this zone was overlain by a high-resistivity and low-permeability zone near the surface. The presence of a sulfur layer or native sulfur veins is likely to have caused this low-permeability zone. Low-resistivity zones were also found around the crater floor. In particular, the location of the low-resistivity zone beneath the southern slope of the inner crater wall was consistent with that of a pressure source. This may indicate the formation of a cap structure that seals the upwelling hydrothermal fluids. At greater depths, the low-resistivity zone extended westward, consistent with the distribution of hydrothermally altered zones and hot springs. Considering the results of water resistivity measurements along the Iwo River, which confirmed hydrothermal fluid seepage at several locations, the hydrothermal system beneath NC could be continuous to the west. High-resistivity bodies, interpreted as unaltered rocks, were found in the northern and southern parts of NC, suggesting that these rocks constrained the past development of the hydrothermal system of NC. Graphical Abstract

ABSTRACT Scaling relationships for earthquake source parameters provide fundamental information for characterizing source faults and simulating ground motions and/or tsunamis for scenario earthquakes. For subduction earthquakes, although variations in the physical properties of subducting oceanic plates, such as temperature, subduction velocity, and seismogenic zone depth, have been identified between subduction zones with different tectonic environments, whether variations in source-scaling relationships exist is still under debate. We investigated the regional variations in source parameters for subduction earthquakes by extracting these parameters from heterogeneous slip models for many events (Mw 6.7–9.1) in nine regions, based on source inversion analyses conducted by the U.S. Geological Survey. Regional differences were observed in the mean residuals of the extracted rupture areas with respect to the scaling relationship, although these differences were not substantial enough to reach statistical significance. Meanwhile, by investigating the depth of the source fault for each event, we found a depth-related variation indicating that the size of the rupture area correlates with the top depth of the source fault. This correlation leads to a magnitude-dependent scatter in rupture area and complicates the self-similar scaling model for events with Mw less than 8.4. For small magnitude (Mw<7.5), the event for which rupture propagates upward and extends to the shallow part of the subduction zone has a larger rupture area, whereas the event propagating only downward has a smaller rupture area. The coexistence of these two event types for Mw less than 7.5 increases the scatter in the scaling relationship of rupture area. For large magnitude (Mw 7.5–8.4), most events extend ruptures to the shallow part of the subduction zone and have larger rupture areas, and thus the scatter is smaller than that for small Mw. The association between rupture propagation direction and rupture area may be useful for updating seismic hazard assessment for subduction earthquakes.

Abstract Physics-informed neural networks have attracted significant attention in scientific machine learning for their capability to solve forward and inverse problems governed by partial differential equations. However, the accuracy of PINN solutions is often limited by the treatment of boundary conditions. Conventional penalty-based methods, which incorporate boundary conditions as penalty terms in the loss function, cannot guarantee exact satisfaction of the given boundary conditions and are highly sensitive to the choice of penalty parameters. This paper demonstrates that distance functions, specifically R-functions, can be leveraged to enforce boundary conditions, overcoming these limitations. R-functions provide normalized distance fields, enabling flexible representation of boundary geometries, including non-convex domains, and facilitating various types of boundary conditions. Nevertheless, distance functions alone are insufficient for accurate inverse analysis in PINNs. To address this, we propose an integrated framework that combines the normalized distance field with bias-corrected adaptive weight tuning to improve both accuracy and efficiency. Numerical results show that the proposed method provides more accurate and efficient solutions to various inverse problems than penalty-based approaches, even in the presence of non-convex geometries with complex boundary conditions. This approach offers a reliable and efficient framework for inverse analysis using PINNs, with potential applications across a wide range of engineering problems.

Abstract Determining the mechanical properties of thin films presents significant challenges due to their nanometer-scale thickness. The separation of thin films from their substrates for standard plastometric testing is often difficult, if not impossible, complicating the direct measurement of their properties. Consequently, nanoindentation tests, which involve using small indenters and analyzing force-displacement curves, are commonly employed to assess the mechanical properties of thin films. However, experimental methods alone may be insufficient for accurately determining these properties for such thin films. This paper proposes an approach that combines numerical modelling of nanoindentation tests with the finite element method and inverse analysis to determine the optimal material constants for the substrate and thin film. The study focuses on TiN thin films deposited on silicon and stainless steel substrates as case studies. Prior to extracting the properties of the thin films, a comprehensive numerical accuracy analysis of the nanoindentation model was conducted. This involved investigating the impact of the digital model on the accuracy of results, comparing 2D and 3D models to optimize computational efficiency, and analysing the effect of finite element mesh discretization. The critical importance of accurately representing the indenter shape for reliable results was also highlighted. Following model validation, a series of nanoindentation simulations were performed on silicon and subsequently on the TiN/Si structure, enabling the separate determination of material constants for the substrate and the TiN thin film. The procedure was then applied to the TiN/SS structure for verification. The findings demonstrate that this approach enables the determination of the as-deposited thin film material properties based solely on nanoindentation tests and a robust numerical model, and it can be extended to other thin films.

Objective: This study presents a single-center retrospective analysis evaluating the long-term outcomes of patients with non-A non-B aortic dissection who underwent either total arch replacement with frozen elephant trunk (TAR with FET) or thoracic endovascular aortic repair (TEVAR). Methods: From 2010 to 2022, patients with non-A non-B aortic dissection who received TEVAR or TAR with FET were selected for clinical data collection and long-term follow-up. Baseline characteristics were balanced using inverse probability weighting. Result: A total of 186 patients were included, with 123 in the TEVAR group and 63 in the FET group. No significant difference in 30-day mortality was observed between the FET and TEVAR groups in both unadjusted and inverse probability weighting analyses (P > 0.05). The FET group, however, consistently showed higher rates of continuous renal replacement therapy (11.6% vs 0.53%, P＜0.01) and prolonged intensive care unit stay (81.2 ± 70 h vs 38.3 ± 51.7 h, P＜0.01). Kaplan-Meier curve analysis showed no significant difference in overall long-term survival (P > 0.05), but landmark analysis revealed that after 6 years post-operation, the FET group had significantly higher survival rates (P＜0.01). Besides, there was no significant difference in aortic-related reintervention between groups (P > 0.05). Conclusion: Regarding early prognosis, it is difficult to determine the superiority between these two approaches. However, from a long-term perspective, TAR with FET represents a more effective treatment strategy compared to TEVAR for patients with non-A non-B type aortic dissection.

Background: Acute heart failure (AHF) with pre-cardiogenic shock carries substantial morbidity and mortality. Istaroxime, a novel intravenous inotrope that enhances both cardiomyocyte contractility and relaxation without adrenergic overstimulation, has demonstrated favourable haemodynamic and echocardiographic effects in early randomized trials. We sought to update the evidence including the newly reported SEISMiC trial to better define istaroxime’s role in AHF pre-cardiogenic shock. Methods: We performed a systematic search of PubMed, Web of Science, SCOPUS, and the Cochrane Library for RCTs comparing istaroxime versus placebo in adult hospitalized AHF patients with pre-cardiogenic shock (SBP <90 mmHg without hypoperfusion or persistent hypotension 70–100 mmHg for ≥2 h). Data were extracted independently by two reviewers. Continuous endpoints (e.g., SBP change) were pooled using random effects inverse variance meta analysis and reported as mean differences (MD) with 95% confidence intervals (CI); dichotomous outcomes used risk ratios (RR). Heterogeneity was assessed with I 2 . Results: Five studies comprising 360 patients were identified. For change in systolic blood pressure at 24 h, pooled MD was +5.4 mmHg (95% CI 2.6, 8.3; p <0.001; I 2 = 0%), favouring istaroxime. Heart rate decreased by −3.1 bpm (95% CI −5.2, −0.9; p = 0.005; I 2 = 0%). Stroke volume index improved (MD +3.0 mL/m 2 ; 95% CI 2.4, 3.6; p <0.0001; I 2 = 0%). In the SEISMiC trial, istaroxime (1.0–1.5 µg/kg/min) significantly increased SBP area-under-the-curve at 6 h (difference +22.2 mmHg; p = 0.017) and 24 h (difference +82.5 mmHg; p = 0.025) without excess serious adverse events. No significant increase in arrhythmias or myocardial injury markers was observed (RR for worsening HF events 1.2; 95% CI 0.7, 2.1; p = 0.45; I 2 = 5%). Conclusions: Istaroxime demonstrates consistent, clinically meaningful improvements in blood pressure, cardiac output, and diastolic function with a favourable safety profile in AHF pre-cardiogenic shock. The inclusion of the SEISMiC trial strengthens the evidence base, supporting further large scale Phase III investigations.

Multilayer dielectric thin films are fundamental components in modern microelectronic and photonic devices where thermal management is critical. This work presents a robust methodological framework for accurately determining the cross-plane effective thermal conductivity (κeff,⊥) of such complex systems using the 3-omega method. We use a five-layer Si3N4/SiO2 stack fabricated by plasma-enhanced chemical vapor deposition as a case study. The analysis combines experimental data from multiple heater geometries with a 3D finite element method based inverse analysis. We first demonstrate, through a frequency-dependent sensitivity analysis, that a direct multi-parameter fit for intrinsic layer properties is an ill-posed problem. This analysis provides a clear, quantitative justification for adopting a simpler and more robust effective medium model (EMA). The validity and application boundaries of the EMA are then rigorously established through a numerical study on a series of "virtual samples." Finally, applying this validated framework to our experimental data, the thermal conductivity of a fused silica substrate was determined to be 1.287 ± 0.030 W/(m K), and the effective thermal conductivity of the 1288 nm thick stack was reliably determined to be 0.621 ± 0.008 W/(m K). This work provides not only a key thermophysical property for Si3N4/SiO2 multilayers but also a comprehensive and validated workflow for reliably characterizing complex thin film systems where standard analytical solutions fail.

A machine learning-based inverse analysis procedure for concrete softening law prediction using non-experimental datasets

The forward and inverse analysis of tunnel based on weak form physics-informed deep learning

The objective of the present study was to explore the bi-directional causal relationship between IPF and diabetes (type 1 diabetes and type 2 diabetes)/ diabetic nephropathy/glycemic traits [fasting glucose and glycated hemoglobin (HbA1c)]/fasting blood insulin through MR analysis. A bi-directional two-sample Mendelian randomization (MR) study design was adopted to evaluate the causal relationship between IPF and diabetes (type 1 diabetes and type 2 diabetes), diabetic complications (diabetic nephropathy) and glycemic traits [fasting blood glucose, glycated hemoglobin (HbA1c), fasting insulin] in a European population. Genome-wide association study summary data was obtained. The inverse variance weighted (IVW) method with a fixed-effects model was used to estimate the primary causal effects. The causal effects are represented by reporting odds ratios (OR) and their corresponding 95% confidence intervals (CI). The robustness of results was assessed using the MR-Egger and Weighted Median methods. In the forward MR analysis, the IVW method revealed a significant causal effect of IPF on type 2 diabetes (OR=1.031, 95% CI: 1.011-1.052). Similar estimates were obtained through the Weighted Median method. However, no significant causal effects were observed on type 1 diabetes, diabetic nephropathy, fasting blood glucose, HbA1c, and fasting insulin, respectively (p>0.05). We performed the reverse MR analysis using similar methods to the forward MR approach. MR analysis only showed a significant causal association of fasting insulin with IPF risk, with an OR of 3.576 (95% CI: 1.958-6.531). Genetically determined IPF was linked to an elevated risk of type 2 diabetes. The inverse MR analysis indicated that there was no causal impact of genetically predicted type 2 diabetes on the IPF risk. Genetically predicted fasting blood insulin was found to be positively associated with IPF risk.

Satellite-derived bathymetry (SDB) using Ice, Cloud, and Land Elevation satellite-2 (ICESat-2) LiDAR data and remote sensing images faces challenges in the difficulty of uniform coverage of the inversion area by the bathymetric control points due to the linear sampling pattern of ICESat-2. This study proposes a novel virtual control point optimization framework integrating inverse distance weighting (IDW) and spectral confidence analysis (SCA). The methodology first generates baseline bathymetry through semi-empirical band ratio modeling (control group), then extracts virtual control points via SCA. An optimization scheme based on spectral confidence levels is applied to the control group, where high-confidence pixels utilized a residual correction-based strategy, while low-confidence pixels employed IDW interpolation based on a virtual control point. Finally, the preceding optimization scheme uses weighting-based fusion with the control group to generate the final bathymetry map, which is also called the optimized group. Accuracy assessments over the three research areas revealed a significant increase in accuracy from the control group to the optimized group. When compared with in situ data, the determination coefficient (R2), RMSE, MRE, and MAE in the optimized group are better than 0.83, 1.48 m, 12.36%, and 1.22 m, respectively, and all these indicators are better than those in the control group. The key innovation lies in overcoming ICESat-2’s spatial sampling limitation through spectral confidence stratification, which uses SCA to generate virtual control points and IDW to adjust low-confidence pixel values. It is also suggested that when applying ICESat-2 satellite data in active–passive-fused SDB, the distribution of training data in the research zone should be adequately considered.

Summary Nowadays, many joint inversions are carried out to understand the earthquake source process. In the joint inversion analysis, we have to determine the relative weights among different datasets in addition to the regularization term, such as smoothing. Akaike’s Bayesian Information Criterion (ABIC) is known to be useful to find the appropriate values of such hyperparameters. This study proposes a method to jointly invert tsunami, GNSS, and SAR data using ABIC to construct a finite fault model. We demonstrate our inversion scheme in the case of the 2024 Noto Peninsula earthquake, whose fault geometry is still under discussion. Since the dip angle of the fault can also be considered as a hyperparameter, we evaluate three types of dip angles and estimate an appropriate value based on ABIC. In other words, our inversion scheme utilizes ABIC to determine the dip angle, the weights among datasets, and the spatial smoothness of fault slip. Our fault model indicates that (1) listric fault, varying the dip angle with depth, is the most appropriate among the ones we proposed, (2) the largest slip is on the fault under the northwestern corner of the peninsula, and (3) coseismic fault slip extends to offshore faults east of the peninsula. In the case of the listric fault, ABIC values GNSS and SAR data, which improves the agreement of the on-land coseismic displacement while also reproducing tsunami data. We also find that analyzing tsunami records in the frequency domain helps to obtain a robust inversion result when employing ABIC.

The South China Sea is surrounded by many large cities with developed economy and dense population along the coast, where tsunamis are major potential disasters within the sea area. Therefore, a dataset and comprehensive assessment of potential tsunami scenarios are an important foundation for design of coastal and ocean engineering, disaster prevention and mitigation strategy. Based on historical seismic records and geodetic observation data, this study integrates ocean dynamic model, the fault coupling inversion and source uncertainty analysis, to construct a potential tsunami scenario dataset in the South China Sea. The scenario dataset covers all possibilities of tsunamis caused by earthquakes with Mw ≥ 7. Both tsunami height and period have been systematically provided, which has not been given by existing researches. The dataset allows users to obtain the joint probability distribution of tsunami wave height and period at any location of interest in the South China Sea, providing key parameters for hazard prevention and mitigation.

Objective To investigate the effects of fatigue on lower limb kinematics and kinetics during manual lifting tasks and to quantitatively analyze these effects in order to provide guidance for safe work practices. Methods Twenty healthy male college students performed lifting tasks with two load conditions (15 kg, low load; and 25 kg, high load) before and after fatigue. An eight-camera 3D motion capture system and two force plates were used to collect surface marker trajectories and ground reaction force data. Inverse kinematics and inverse dynamics analyses were conducted using OpenSim to calculate movement duration, joint angles, joint angular velocities, joint moments, joint power, and joint energy expenditure. Results (1) For the 15 kg lifting task, there were no significant differences in any parameter between pre- and post-fatigue conditions. (2) For the 25 kg task, compared to the pre-fatigue state, subjects exhibited decreased movement duration, increased joint range of motion, faster angular velocities, and elevated joint power and energy expenditure after fatigue. Conclusion Under low load conditions, the primary kinematic and kinetic parameters of the lower limb joints remained stable before and after fatigue, demonstrating strong fatigue resistance. In contrast, under high-load conditions, fatigue altered the lower limb movement patterns. The combined effect of high load and fatigue not only increased the burden on the musculoskeletal system but also led to a rise in potential injury risk, which requires further research for validation.

Background Cutaneous melanoma (CM) is a highly lethal skin tumor. Some patients respond poorly to existing therapies, and developing new targeted therapies remains challenging. Methods We combined the results of eQTLs, pQTLs, and genome-wide association study (GWAS) to identify potential causal effects of two target genes on CM, based on multi-omics Mendelian randomization (MR). Sensitivity analysis, co-localization analysis, and inverse MR analysis were also employed to verify the robustness of this causal relationship. Multi-omics data were then applied to explore the expression patterns of immune infiltration of the target genes and construct nomogram models. Results The results showed that the gene prediction levels of EPS15L1 and HGS were associated with an increased risk of CM. Co-localization analysis revealed significant horizontal pleiotropy of the target gene, and reverse MR showed unidirectional causality of the targets. Multi-omics analysis comprehensively demonstrated the expression regulation pattern of the target genes in the CM immune-environment and identified interactions between EPS15L1 (Q9UBC2) and HGS (O14964) and doxorubicin, demonstrating the potential for drug application. The validity of the targets was further verified by molecular biology experiments. Conclusion This study provides robust genetic and therapeutic evidence for targeting EPS15L1 and HGS in CM treatment.

In this study, numerical simulations were performed, and leaks in viscoelastic pipelines were detected. Based on the transient flow equations derived from the continuity and momentum equations, the Kelvin–Voigt model was used to describe the viscoelastic constitutive relationship and derive the strain equation, further establishing a one-dimensional transient flow model for viscoelastic pipelines. A frequency-domain analysis of the transient flow was performed by deriving the Fourier transform and transfer matrix. An inverse problem analysis method for transient flow leak detection was proposed to identify the leak location and rate by minimizing the objective function. To verify the effectiveness of the proposed model, an experimental platform was built, and the pressure head frequency-domain data under working conditions of no leak, experimental leak, and simulated leak were compared. The results showed that the experimental data were consistent with the simulated data under leakage conditions, thus proving that the model was accurate and reliable. Under leak-free conditions, the frequency-domain characteristics of transient pressure waves exhibit significant symmetrical features, whereas when a leak exists in the pipeline, the leak point acts as a localized non-uniform disturbance source, disrupting the symmetry of the frequency-domain characteristics. Moreover, the leak point can be determined by the difference in the peak heights between the no-leak and leak conditions, and the leak parameters can be accurately identified using the inverse problem method.

In this study, we develop a conditional diffusion model that proposes the optimal process parameters and predicts the microstructure for the desired mechanical properties. In materials development, it is costly to try many samples with different parameters in experiments and numerical simulations. The use of data-driven inverse design method can reduce the cost of materials development. This study develops an inverse analysis model that predicts process parameters and microstructures. This method can be used for any material, but in this study it is applied to polymeric material, which is the matrix resin of carbon fiber reinforced thermoplastics as an example. Matrix resins contain a mixture of dendrites, which are crystalline phases, and amorphous phases even after crystal growth is complete, and it is important to consider the microstructures consisting of the crystalline structure and the remaining amorphous phase to achieve the desired mechanical properties. Typically, the temperature during forming affects the microstructures, which in turn affect the macroscopic mechanical properties. The trained diffusion model can propose not only the processing temperature but also the microstructure when Young’s modulus and Poisson’s ratio are given. The capability of our conditional diffusion model to represent complex dendrites is also noteworthy. This model can be applied to other process parameters and mechanical properties. Furthermore, multiple process parameters and mechanical properties can be handled together.

Stress and Inverse Analyses of Linear Orthotropic Elastic Materials