Modeling Approach Research Articles

A machine learning-guided modeling approach to the kinetics of α-tocopherol and myricetin synergism in bulk oil oxidation

The shelf-life and quality of food products depend heavily on antioxidants, which protect lipids from free radical degradation. α-Tocopherol and myricetin, two potent antioxidants, synergistically enhance the prevention of oxidative rancidity in bulk oil systems. Understanding their degradation kinetics is essential for deepening our knowledge of their mechanisms and developing strategies to predict shelf-life before expiration. This paper introduces a generalized mathematical model to describe the degradation kinetics of α-tocopherol in the presence of myricetin. Using direct differential methods guided by a machine learning approach based on neural differential equations, we uncover two distinct phases of α-tocopherol degradation when coexisting with myricetin at varying concentration ratios. These findings inform the development of a mixed Weibull model that accurately captures the degradation process. Our study enhances the understanding of antioxidant interactions and provides a reliable method for predicting food system stability, offering valuable insights for optimizing natural antioxidants in food preservation.

Capturing sclera anisotropy using direct collagen fiber models. Linking microstructure to macroscopic mechanical properties.

Because of the crucial role of collagen fibers on soft tissue mechanics, there is great interest in techniques to incorporate them in computational models. Recently we introduced a direct fiber modeling approach for sclera based on representing the long-interwoven fibers. Our method differs from the conventional continuum approach to modeling sclera that homogenizes the fibers and describes them as statistical distributions for each element. At large scale our method captured gross collagen fiber bundle architecture from histology and experimental intraocular pressure-induced deformations. At small scale, a direct fiber model of a sclera sample reproduced equi-biaxial experimental behavior from the literature. In this study our goal was a much more challenging task for the direct fiber modeling: to capture specimen-specific 3D fiber architecture and anisotropic mechanics of four sclera samples tested under equibiaxial and four non-equibiaxial loadings. Samples of sclera from three eyes were isolated and tested in five biaxial loadings following an approach previously reported. Using microstructural architecture from polarized light microscopy we then created specimen-specific direct fiber models. Model fiber orientations agreed well with the histological information (adjusted R2's>0.89). Through an inverse-fitting process we determined model characteristics, including specimen-specific fiber mechanical properties to match equibiaxial loading. Interestingly, the equibiaxial properties also reproduced all the non-equibiaxial behaviors. These results indicate that the direct fiber modeling method naturally accounted for tissue anisotropy within its fiber structure. Direct fiber modeling is therefore a promising approach to understand how macroscopic behavior arises from microstructure.

Multiscale Analysis of Impact-Resistance in Self-Healing Poly(ethylene-co-methacrylic acid) (EMAA) Plain Woven Composites

A study combining multiscale numerical simulation and low-velocity impact (LVI) experiments was performed to explore the comprehensive effects on the impact-resistance of EMAA filaments incorporated as thermoplastic healing agents into a plain woven composite. A multiscale micro–meso–macro modeling framework was established, sequentially propagating mechanical performance parameters among micro–meso–macro models. The equivalent mechanical parameters of the carbon fiber bundles were predicted based on the microscopic model. The mesoscopic representative volume element (RVE) model was crafted by extracting the actual architecture of the monolayer EMAA filaments encompassing the plain woven composite. Subsequently, the fiber and matrix of the mesoscopic model were transformed into a monolayer-equivalent cross-panel model containing monolayers aligned at 0° and 90° by local homogenization, which was extended into a macroscopic equivalent model to study the impact-resistance behavior. The predicted force–time curves, energy–time curves, and damage profile align closely with experimental measurements, confirming the reliability of the proposed multiscale modeling approach. The multiscale analysis reveals that the EMAA stitching network can effectively improve the impact-resistance of plain woven composite laminates. Furthermore, there exist positive correlations between EMAA content and both impact-resistance and self-healing efficiency, achieving a self-healing efficiency of up to 98.28%.

A structural modeling approach to determine the effects of orthorexia nervosa and social appearance anxiety on healthy lifestyle behaviors in young adults

Aims: The aim of this study is to examine the relationship between orthorexia nervosa (ON) and social appearance anxiety (SAA) and to investigate the effects of these variables on healthy lifestyle behaviors (HLB) in young adults aged 18-30 years. Methods: A total of 408 young adult male and female volunteers, with a mean age of 21.7±2.95 years, participated in this study. A structural model was proposed to examine the effects of SAA and ON on HLB. To assess how well the study variables represented the underlying components, we first evaluated a confirmatory measurement model. Subsequently, we analyzed a structural model that allowed for the examination of multiple regression equations involving several dependent variables. Results: SAA had a negative influence on HLB (std β=-0.673; t-value=-8.425), a significant negative relationship between ON, HLB was also promoted (std β=-0.554; t-value=-9.388). It was determined that there is a positive correlation between SAA, ON which is statistically significant (correlation coefficient=0.830; t-value=10.244). Conclusion: It is concluded that the individuals with high social appearance anxiety do not adopt healthy lifestyle behaviors and are more prone to orthorexia nervosa.

A Modeling Approach for Cargo Transportation Considering Energy Saving

This study presents a cargo transportation system management model that enables users to carry out the cargo process efficiently and economically. Cargo transportation is an important part of the transportation systems network. The advantages of cargo transportation are to ensure the safe delivery of products, save time and reduce costs. This study addresses the solution to the problems of high carbon dioxide emissions, and late and expensive deliveries caused by large distribution networks, by taking into account the density and logistics strategies of the cargo companies. The proposed cargo management system focuses on delivering cargo to destinations along the route that users are currently traveling on. This system is built around three main components: optimized route planning by an A* algorithm, a system model controllable through mobile and web interfaces, and nodes designated as drop-off and pick-up points for cargo. The A* algorithm runs with a reward matrix that also takes direction into account to calculate the optimal route. Users carry out the cargo management and tracking processes on mobile and web interfaces. Automatic nodes, which are the model prototype of the study, represent the cargo vending machines where users deliver/receive their cargo. Through this work, users can view and select cargo to carry while traveling between destinations and earn profits by acting as cargo carriers/drivers.

Evaluating land use and climate change impacts on Ravi river flows using GIS and hydrological modeling approach

The study investigates the interplay of land use dynamics and climate change on the hydrological regime of the Ravi River using a comprehensive approach integrating Geographic Information System (GIS), remote sensing, and hydrological modeling at the catchment scale. Employing the Soil and Water Assessment Tool (SWAT) model, simulations were conducted to evaluate the hydrological response of the Ravi River to both current conditions and projected future scenarios of land use and climate change. This study differs from previous ones by simulating future land use and climate scenarios, offering a solid framework for understanding their impact on river flow dynamics. Model calibration and validation were performed for distinct periods (1999–2002 and 2003–2005), yielding satisfactory performance indicators (NSE, R2, PBIAS = 0.85, 0.83, and 10.01 in calibration and 0.87, 0.89, and 7.2 in validation). Through supervised classification techniques on Landsat imagery and TerrSet modeling, current and future land use maps were generated, revealing a notable increase in built-up areas from 1990 to 2020 and projections indicating further expansion by 31.7% from 2020 to 2100. Climate change projections under different socioeconomic pathways (SSP2 and SSP5) were derived for precipitation and temperature, with statistical downscaling applied using the CMhyd model. Results suggest substantial increases in precipitation (10.9 − 14.9%) and temperature (12.2 − 15.9%) across the SSP scenarios by the end of the century. Two scenarios, considering future climate conditions with current and future land use patterns, were analyzed to understand their combined impact on hydrological responses. In both scenarios, inflows to the Ravi River are projected to rise significantly (19.4 − 28.4%) from 2016 to 2100, indicating a considerable alteration in seasonal flow patterns. Additionally, historical data indicate a concerning trend of annual groundwater depth decline (0.8 m/year) from 1996 to 2020, attributed to land use and climate changes. The findings underscore the urgency for planners and managers to incorporate climate and land cover considerations into their strategies, given the potential implications for water resource management and environmental sustainability.

Numerical modeling of heat transfer mechanism in remote sensing bulb of thermal expansion valves

Thermal management of automotive systems has garnered a lot of focus in recent times to improve the energy efficiency of vehicles and address the challenge of global warming. In combustible fuel vehicles, the heat pump compressor is driven by the engine power which leads to changes in operating condition of the cycle while the vehicle is in motion. In electric vehicles, a combined thermal management strategy handles both cabin and battery pack cooling/heating depending on the ambient conditions and power requirements by adjusting refrigerant flow rates and flow directions with sophisticated valves. Irrespective of the vehicle type, effective variable refrigerant flow control is prerequisite to ensure optimum thermal comfort with minimum energy. Thermal expansion valves (TXV) are the most widely used method for regulating mass flow rate of refrigerant during the expansion step of the vapor compression cycle based on the principle of trying to maintain a fixed degree of superheat at the evaporator outlet. Conventional models of the thermal expansion valve ignore the time lag between fluctuation of temperature at the evaporator outlet and the corresponding change in pressure of the TXV remote sensing bulb, treating the valve as an instantaneous element during transient operation. This paper presents a novel method for determining the temperature profile and time constant of the remote sensing bulb response using numerical methods, in an attempt to capture the dynamics of thermal expansion valve in active heat pump cycle. Finite difference method was used to solve for the conduction heat transfer between the bulb and the heat exchanger outlet tube in perfect thermal contact applying the necessary boundary conditions. This improved modeling approach will be helpful in better prediction of the dynamic behavior of thermal expansion valves and how it influences the evaporative heat exchanger performance during transient operations to determine control algorithm and strategies.

Unveiling the impact of climate diversity on solar PV soiling

ABSTRACT Advanced knowledge of PV soiling can aid in the economic profitability of PV plants and efficient generation. This study introduces a novel approach for evaluating energy losses from soiling deposition on PV modules. Experimental investigations conducted under real outdoor conditions in Jaipur (India) were used to validate the model’s accuracy, demonstrating close agreement (with MAPE < 3.6%). Annual soiling trends were also investigated for seven cities reflecting diverse climatic zones (hot-dry, warm-humid, composite, temperate, and cold) and pollutant (PM10) concentrations. The proposed model was also tested with three different cleaning thresholds using rainfall intensities of 1, 3, and 5 mm/day and energy yield reductions due to soiling were simulated. Consequently, the findings of this study underscore the significance of seasonal variations in soiling profiles for different cities and, thus, highlight the need to develop location-specific cleaning strategies for PV systems that also account for the soiling removal due to rain events. The significant contribution of this research is a simplistic modeling approach that addresses the relationship of PV soiling with the input parameters such as wind speed (WS), relative humidity (RH), rainfall, particulate matter (PM10), and tilt angle. Hence, the proposed method can calculate soiling losses worldwide irrespective of geographical differences.

Coupling stress and transmissivity to define equivalent directional hydraulic conductivity of fractured rocks

A DFN (Discrete Fracture Network) modelling approach is developed to couple stresses with fracture transmissivities and to evaluate large scale rock mass hydraulic conductivity. The transmissivity-stress coupling relies on a negative exponential correlation between normal stress acting on a fracture and fracture transmissivity, bounded by residual and maximal apertures. The remote stresses and the local stress fluctuations induced by the fractures themselves are combined in a semi-analytical approach to compute the normal stress acting on each fracture of a DFN. Directional equivalent hydraulic conductivities are numerically computed in all directions from a spherical permeameter setup. The resulting properties are first a cloud of points, where each point defines a direction and an equivalent hydraulic conductivity. The distribution of equivalent hydraulic conductivities is analyzed to define mean values, preferential directions and anisotropy ratio. The entire workflow is developed in the numerical platform DFN.lab. The capacity of the method to investigate the impact of the in-situ stresses on the rock mass hydraulic conductivity is illustrated for fracturing and in-situ stress conditions similar to the conditions at the Forsmark site in Sweden. We find that the stress fluctuations induced by the fractures have a significant impact on the resulting hydraulic conductivity field. They limit the anisotropy ratio to values close to a factor of 3 while the transmissivity distribution is correlated to orientations and spans several orders of magnitude. Sensitivity analyses, performed by changing the parameters of the transmissivity-stress law, show quantitatively how the directional hydraulic conductivities are rather controlled by the orientations of the in-situ stresses or by the underlying connectivity structure of the DFN.

A Monte-Carlo approach for crack initiation modeling of cast superalloys informed by crystal plasticity

Crystal plasticity simulations of materials to assess the fatigue response on the microscale are becoming increasingly popular in industrial application. However, in the case of parts made from cast Ni-base superalloys, the amount of pores in a part combined with the fine spatial discretization around the pores needed due to strong mechanical gradients quickly leads to a computationally impractical number of mesh elements. In this paper we show an alternative to explicit crystal plasticity modeling of part-scale porosity by introducing a Monte-Carlo submodel that recombines the fatigue response of single pores predicted by crystal plasticity into the fatigue response of the pore agglomerate. The model can be applied in realtime due to the use of precomputed crystal plasticity results. We demonstrate that fatigue indicator parameters predicted by the Monte-Carlo submodel agree well with those predicted by explicit crystal plasticity simulations. Lastly, we apply the proposed model to study the influence of different porosity volume percentages, pore sizes and pore morphologies on the fatigue indicator parameters of a cast MAR-M247 superalloy.

Distributed Space Team Coordination: An Agent-based Modeling and Simulation Approach

Distributed Space Team Coordination: An Agent-based Modeling and Simulation Approach

Circular economy and critical barriers: Mapping the pathways and success metrics for sustainable circular success in industrialised South Asian developing nations

The economy of a developing country is often industry centric, and is recognised globally. The circular economy (CE) model involves a manufacturing approach where products are systematically remanufactured and recycled within facilities to minimize waste. The emerging domain of CE in industrialised South Asian economies offers considerable benefits for developing countries due to its sustainable production, although its adoption is obstructed by various barriers. This research aims to develop a model for adopting CE within the industrial sector of South Asian developing nations by analysing the interrelation between overcoming circular economy adoption barriers (CEABs) and sustainable circular success (SCS). To attain this, questionnaire was completed by 310 industrial experts in Pakistan. Furthermore, a partial least square structural equation modelling (PLS-SEM) approach was employed to specify the barriers and inspect the interrelation between overcoming CEABs and SCS. The findings revealed a high correlation, with addressing the CEABs contributing 66.1 % to the SCS of the industrial sector. The results of this study indicate the outer loadings and average variance extracted values for all constructs surpassed the minimum threshold of 0.5, validating their acceptance. Moreover, the constructs in the study were measured reliably, exceeding a value of 0.8, which signify strong internal consistency. Additionally, the average path coefficient value (β) of 0.172 indicates a medium and positive correlation between CEABs and SCS. The study's findings can be used as a reference for policymakers to explore the primary barriers to CE adoption in developing nations and achieve SCS in industrial projects.

Modelling permanent strain behaviour of soil for flexible pavement design utilizing LWD measurements

ABSTRACT This study aims to improve understanding of unbound material behavior by modeling permanent strain using Light Weight Deflectometer (LWD) measurements, as part of VTI’s broader LWD research. By comparing permanent strain modelling from LWD data with Repeated Load Triaxial (RLT) tests on silty sand soils, it validates LWD data accuracy and explores material property influences on permanent strain parameters, revealing a clear correlation with water content and stress levels. Using LWD data validated against RLT tests, the study delves into material behaviour understanding under varied conditions, focusing on comparing model parameters of permanent strain prediction derived from LWD and RLT tests. Despite minor differences, both methods produce acceptable values, affirming the modelling approach’s robustness. The findings demonstrate LWD’s potential to improve pavement design by providing insights into permanent strain behavior that supports mechanistic-empirical approaches, leading to more efficient designs and better resource use for improved performance.

NPCoronaPredict: A Computational Pipeline for the Prediction of the Nanoparticle-Biomolecule Corona.

The corona of a nanoparticle immersed in a biological fluid is of key importance to its eventual fate and bioactivity in the environment or inside live tissues. It is critical to have insight into both the underlying bionano interactions and the corona composition to ensure biocompatibility of novel engineered nanomaterials. A prediction of these properties in silico requires the successful spanning of multiple orders of magnitude of both time and physical dimensions to produce results in a reasonable amount of time, necessitating the development of a multiscale modeling approach. Here, we present the NPCoronaPredict open-source software package: a suite of software tools to enable this prediction for complex multicomponent nanomaterials in essentially arbitrary biological fluids, or more generally any medium containing organic molecules. The package integrates several recent physics-based computational models and a library of both physics-based and data-driven parametrizations for nanomaterials and organic molecules. We describe the underlying theoretical background and the package functionality from the design of multicomponent NPs through to the evaluation of the corona.

A unified modeling approach for concrete masonry shear walls: Conception, validation and insights

Distinct modeling strategies have been developed to simulate concrete masonry shear walls; however, each is validated with a limited number of specimens tested under identical laboratory conditions and only covers specific detailing scenarios. This research seeks a unified finite element-based modeling approach integrating macro-modeling concepts for grouted masonry (GM) regions and simplified micro-modeling for ungrouted masonry (UGM) regions. The objective is to facilitate accurate simulation of the in-plane behavior of diverse typologies of concrete masonry shear walls: unreinforced masonry (URM), fully grouted reinforced masonry (FGRM), and partially grouted reinforced masonry (PGRM). A smeared crack approach is used to capture the crushing and cracking of grouted masonry (GM) and ungrouted masonry (UGM), and interface elements are incorporated into UGM regions to model the mortar joints, resulting in a combined smeared-discrete model. The validation process comprises two levels. The first-level evaluation focuses on URM and FGRM walls, emphasizing the analysis and comparison of lateral force–displacement curves. Second-level validation is done through simulation of PGRM walls by evaluating lateral force–displacement curves, damage patterns, and reinforcement strains. The implemented modeling approach generally simulates the response of the different experimental masonry shear walls evaluated with good accuracy. A sensitivity analysis was also performed to reveal the influence of some parameters on the numerical results, providing additional perceptions into the reliability of the modeling approach. Relevant insights into the modeling approach were obtained, allowing complex phenomena in the tested walls to be simulated, thus contributing to advances in the simulation of in-plane loaded concrete masonry walls.

The Strength of Bisymmetric Modes in SDSS-IV/MaNGA Barred Galaxy Kinematics

The Sloan Digital Sky Survey-IV/Mapping Nearby Galaxies at Apache Point Observatory (MaNGA) Survey data provide an unprecedented opportunity to study the internal motions of galaxies and, in particular, represent the largest sample of barred galaxy kinematic maps obtained to date. We present results from Nirvana, our nonaxisymmetric kinematic modeling code built with a physically motivated Bayesian forward modeling approach, which decomposes MaNGA velocity fields into first- and second-order radial and tangential rotational modes in a generalized and minimally supervised fashion. We use Nirvana to produce models and rotation curves for 1263 unique barred MaNGA galaxies and a matched unbarred control sample. We present our modeling approach, tests of its efficacy, and validation against existing visual bar classifications. Nirvana finds elevated noncircular motions in galaxies identified as bars in imaging, and bar position angles that agree well with visual measurements. The Nirvana-MaNGA barred and control samples provide a new opportunity for studying the influence of nonaxisymmetric internal disk kinematics in a large statistical sample.

Predictors of Sustainable Tourism Development during the Post-Pandemic Period in Bangladesh

The goal of this study was to examine the predictors of sustainable tourism development during the post-COVID-19 pandemic period in the context of Bangladesh. Quantitative-type research was applied, and this study used a descriptive research design. A standardized questionnaire was used to collect 302 data points from actual indigenous tourists using an online purposive sampling method. A partial least squares structural equation modeling (PLS-SEM) approach was used to evaluate the data and test the hypotheses. The results of the PLS-SEM analysis method demonstrate that environmental integrity, social equity, economic prosperity, and technological adaption had a positive relationship with sustainable tourism development during the post-COVID-19 pandemic period in Bangladesh. This research article provides practical guidelines for tourism authorities (BTB, BPC, and so on) and stakeholders on how to effectively impact environmental integrity, social equity, economic prosperity, and technological adaption on sustainable tourism development during the post-COVID-19 pandemic period in the context of Bangladesh. This study recommends that influential strategic factors be utilized to develop a sustainable tourism business in Bangladesh, positioning it as an emerging tourism destination.

The Effect of Hydrate Formation Conditions on the Mechanics of Laboratory Methane Hydrate‐Bearing Sediments

AbstractThe mechanics of methane hydrate‐bearing sediments (MHBS) have been broadly investigated over recent years in the context of methane‐gas production or climate‐change effects. Their mechanical investigation has mainly been carried out using models constructed from experimental data obtained for laboratory‐formed MHBS. Along with the dominant effects of hydrate saturation and morphology within the host soil pores, this study recognizes the effective pressure at which the hydrate is formed as a key factor in the MHBS mechanics. A state‐of‐the‐art experimental study has been conducted in order to isolate the effect of the hydrate formation pressure, for use as a model parameter. Two generalized mechanical prediction models that incorporate the effect of the hydrate formation pressure are developed in this work: (a) an analytical shear strength prediction, and (b) an empiric graphical model for predicting volumetric changes along a given stress path. The models are related to a novel data representation which enables the analysis of a few individual test outcomes as a whole, through a volume‐change mapping that describes the complex influence of the volumetric effect of hydrate in MHBS, under combined hydrostatic and deviatoric loading scenarios. In this study, we delve into a specific configuration of hydrate morphology, hydrate saturation, and host soil type, enabling a distinctive fundamental geotechnical investigation and the development of a conceptual modeling approach. The paper describes the approaches by which the MHBS properties can be extracted for other MHBS samples (than those examined in this work) having different host soils and hydrate pore‐space morphologies.

Leap, learn, earn: exploring academic risk taking and learning success across gender and socioeconomic groups

Abstract Background The positive learning effects of academic risk taking (ART) in higher education has been discussed since the 1980s. However, this may not apply equally for all social groups. Men and women may differ in the way they use ART to construct their gender identity. Students with different socioeconomic status (SES) may differ in their ability to navigate academic risks due to differences in available cultural capital. Aims This study examines gender and SES disparities in ART and their impact on learning success. It explores if ART mediates and is moderated by gender and SES effects. Additionally, it assesses if ART directly predicts learning success. Sample A sample of N = 381 German university students was used. Methods Data was analyzed following a structural equation modeling approach. Results Men show more ART on the seminar group dimension, whereas women show more ART on the peer dimension. Being male indirectly predicts higher learning success via the seminar group dimension of ART. Furthermore, SES and gender moderate the effect between ART and learning success. Both ART dimensions directly predict students’ learning success. Conclusions Our research contributes to understanding the mechanisms of social disparities within higher education and offers implications for the development of inclusive teaching strategies and research on aspects of intersectionality.

Automated speech analysis for risk detection of depression, anxiety, insomnia, and fatigue: Algorithm Development and Validation Study.

While speech analysis holds promise for mental health assessment, research often focuses on single symptoms, despite symptom co-occurrences and interactions. In addition, predictive models in mental health do not properly assess speech-based systems' limitations, such as uncertainty, or fairness for a safe clinical deployment. We investigated the predictive potential of mobile-collected speech data for detecting and estimating depression, anxiety, fatigue, and insomnia, focusing on other factors than mere accuracy, in the general population. We included n=865 healthy adults and recorded their answers regarding their perceived mental and sleep states. We asked how they felt and if they had slept well lately. Clinically validated questionnaires measuring depression, anxiety, insomnia, and fatigue severity were also used. We developed a novel speech and machine learning pipeline involving voice activity detection, feature extraction, and model training. We automatically analyzed participants' speech with a fully ML automatic pipeline to capture speech variability. Then, we modelled speech with pretrained deep learning models that were pre-trained on a large open free database and we selected the best one on the validation set. Based on the best speech modelling approach, we evaluated clinical threshold detection, individual score prediction, model uncertainty estimation, and performance fairness across demographics (age, sex, education). We employed a train-validation-test split for all evaluations: to develop our models, select the best ones and assess the generalizability of held-out data. The best model was WhisperM with a max pooling, and oversampling method. Our methods achieved good detection performance for all symptoms, depression (PHQ-9 AUC= 0.76F1=0.49, BDI AUC=0.78, F1=0,65), anxiety (GAD-7 F1=0.50, AUC=0.77) insomnia (AIS AUC=0.73, F1=0.62), and fatigue (MFI Total Score F1=0.88, AUC=0.68). These strengths were maintained for depression detection with BDI and Fatigue for abstention rates for uncertain cases (Risk-Coverage AUCs < 0.4). Individual symptom scores were predicted with good accuracy (Correlations were all significant, with Pearson strengths between 0.31 and 0.49). Fairness analysis revealed that models were consistent for sex (average Disparity Ratio (DR) = 0.86), to a lesser extent for education level (average Disparity Ratio (DR) = 0.47) and worse for age groups (average Disparity Ratio (DR) = 0.33). This study demonstrates the potential of speech-based systems for multifaceted mental health assessment in the general population, not only for detecting clinical thresholds but also for estimating their severity. Addressing fairness and incorporating uncertainty estimation with selective classification are key contributions that can enhance the clinical utility and responsible implementation of such systems. This approach offers promise for more accurate and nuanced mental health assessments, benefiting both patients and clinicians.