Binary Response Research Articles

An efficient approach for identifying important biomarkers for biomedical diagnosis

In this paper, we explore the challenges associated with biomarker identification for diagnosis purpose in biomedical experiments, and propose a novel approach to handle the above challenging scenario via the generalization of the Dantzig selector. To improve the efficiency of the regularization method, we introduce a transformation from an inherent nonlinear programming due to its nonlinear link function into a linear programming framework under a reasonable assumption on the logistic probability range. We illustrate the use of our method on an experiment with binary response, showing superior performance on biomarker identification studies when compared to their conventional analysis. Our proposed method does not merely serve as a variable/biomarker selection tool, its ranking of variable importance provides valuable reference information for practitioners to reach informed decisions regarding the prioritization of factors for further investigations.

Self-powered SnSx/TiO2 photodetectors (PDs) with dual-band binary response and the applications in imaging and light-encrypted logic gates

In this work, we report the design and fabrication of self-powered binary response PDs based on II-type heterostructures consisting of SnSx nanoflakes (NFs) and rutile TiO2 nanorod arrays (NRs). The TiO2 NRs effectively block light with wavelengths below 400 nm from reaching SnSx. Under 385 nm light, the photoelectrons in TiO2 recombine with holes in SnSx at the interface due to the energy band bending, resulting in a positive photocurrent. Under 410 nm light, the photoelectrons in SnSx and the photogenerated holes in TiO2 accumulate at the interface, overcoming the interfacial potential barriers induced by the higher Fermi levels of SnSx and inducing a negative photocurrent. Based on the bipolar response, the dual-band imaging capability without external filters and the light-encrypted OR, AND, and NOT logic gates using a single device are demonstrated. This work provides a blueprint for the development of multifunctional self-powered PDs that can simplify system architecture, reduce the energy consumption, and improve accuracy for applications, such as visual systems, light-controlled logic circuits, and encrypted optical communications.

Access to technology, internet usage, and online health information-seeking behaviors in a racially diverse, lower-income population.

This study examined access to technology, internet usage, and online health information-seeking behaviors, in a racially diverse, lower-income population. Data were obtained via a cross-sectional survey of low-income communities in Houston, Los Angeles, and New York between April and August 2023. Binary responses to the following online health information-seeking behaviors, internet and technology access, were examined: using the internet to (i) understand a medical diagnosis, (ii) fill a prescription, (iii) schedule a healthcare appointment, (iv) email communication with a healthcare provider, and (v) access electronic health records and medical notes. 41% of survey respondents identified as non-Hispanic Black individuals, 33% as non-Hispanic White individuals, and 22% as Hispanic individuals. 69% reported a pre-tax annual household income of less than $35,000. 97% reported ownership/access to a smart device; 97% reported access to reliable internet. In the past year, only 59% reported using the internet to better understand their medical diagnosis, 36% reported filling a prescription online, 47% scheduled a medical appointment online, 47% viewed electronic health records online, and 56% emailed healthcare providers. Female sex, higher incomes, and having at least a bachelor's degree were significantly associated with all five online health information-seeking attributes. Despite high technology adoption rates, we observed suboptimal online health information-seeking behaviors. This underutilization has potential adverse implications for healthcare access and use given the documented advantage of HIT. Efforts to increase health information-seeking behaviors should explore the identification of HIT barriers, and patient education to increase familiarity and usage in this population.

Cooperative control of a DNA origami force sensor

Biomolecular systems are dependent on a complex interplay of forces. Modern force spectroscopy techniques provide means of interrogating these forces, but they are not optimized for studies in constrained environments as they require attachment to micron-scale probes such as beads or cantilevers. Nanomechanical devices are a promising alternative, but this requires versatile designs that can be tuned to respond to a wide range of forces. We investigate the properties of a nanoscale force sensitive DNA origami device which is highly customizable in geometry, functionalization, and mechanical properties. The device, referred to as the NanoDyn, has a binary (open or closed) response to an applied force by undergoing a reversible structural transition. The transition force is tuned with minor alterations of 1 to 3 DNA oligonucleotides and spans tens of picoNewtons (pN). The DNA oligonucleotide design parameters also strongly influence the efficiency of resetting the initial state, with higher stability devices (≳10 pN) resetting more reliably during repeated force-loading cycles. Finally, we show the opening force is tunable in real time by adding a single DNA oligonucleotide. These results establish the potential of the NanoDyn as a versatile force sensor and provide fundamental insights into how design parameters modulate mechanical and dynamic properties.

A Spectral Method for Identifiable Grade of Membership Analysis with Binary Responses.

Grade of membership (GoM) models are popular individual-level mixture models for multivariate categorical data. GoM allows each subject to have mixed memberships in multiple extreme latent profiles. Therefore, GoM models have a richer modeling capacity than latent class models that restrict each subject to belong to a single profile. The flexibility of GoM comes at the cost of more challenging identifiability and estimation problems. In this work, we propose a singular value decomposition (SVD)-based spectral approach to GoM analysis with multivariate binary responses. Our approach hinges on the observation that the expectation of the data matrix has a low-rank decomposition under a GoM model. For identifiability, we develop sufficient and almost necessary conditions for a notion of expectation identifiability. For estimation, we extract only a few leading singular vectors of the observed data matrix and exploit the simplex geometry of these vectors to estimate the mixed membership scores and other parameters. We also establish the consistency of our estimator in the double-asymptotic regime where both the number of subjects and the number of items grow to infinity. Our spectral method has a huge computational advantage over Bayesian or likelihood-based methods and is scalable to large-scale and high-dimensional data. Extensive simulation studies demonstrate the superior efficiency and accuracy of our method. We also illustrate our method by applying it to a personality test dataset.

Heterogeneous treatment effect estimation for observational data using model-based forests.

The estimation of heterogeneous treatment effects has attracted considerable interest in many disciplines, most prominently in medicine and economics. Contemporary research has so far primarily focused on continuous and binary responses where heterogeneous treatment effects are traditionally estimated by a linear model, which allows the estimation of constant or heterogeneous effects even under certain model misspecifications. More complex models for survival, count, or ordinal outcomes require stricter assumptions to reliably estimate the treatment effect. Most importantly, the noncollapsibility issue necessitates the joint estimation of treatment and prognostic effects. Model-based forests allow simultaneous estimation of covariate-dependent treatment and prognostic effects, but only for randomized trials. In this paper, we propose modifications to model-based forests to address the confounding issue in observational data. In particular, we evaluate an orthogonalization strategy originally proposed by Robinson (1988, Econometrica) in the context of model-based forests targeting heterogeneous treatment effect estimation in generalized linear models and transformation models. We found that this strategy reduces confounding effects in a simulated study with various outcome distributions. We demonstrate the practical aspects of heterogeneous treatment effect estimation for survival and ordinal outcomes by an assessment of the potentially heterogeneous effect of Riluzole on the progress of Amyotrophic Lateral Sclerosis.

Equity in Digital Health: Assessing Access and Utilization of Remote Patient Monitoring, Medical Apps, and Wearables in Underserved Communities.

This study examined access to, and use of remote patient monitoring (RPM), medical applications, and wearables in a racially diverse, lower-income population. Data were obtained via a cross-sectional survey of adults from low-income communities in Houston, Los Angeles, and New York between April and August 2023. The survey examined access to, and use of RPM, medical applications, and wearables, among respondents. Binary responses to the following questions were examined using logistic regression models: In the past 12 months, have you (i) used RPM, (ii) used a medical app, and (iii) used an electronic wearable device to monitor or track health or activity? A total of 305 surveys were returned, of which 212 were complete (69.5% completion rate). Demographically, 22% self-identified as Hispanic, 41% as non-Hispanic Black individuals, and 33% as non-Hispanic White individuals. Overall, 69% of respondents reported a pre-tax annual household income of less than $35 000 and 96% indicated they own a smart phone. However, only 3 of 10 reported using RPM, 15% reported using a medical app, and 14% reported using wearables. Race was strongly associated with RPM usage, with Black respondents significantly less likely to have used RPM, compared to their white counterparts (OR: 0.31, P = .002). Education (bachelor's degree or more OR: 4.79, P = .03) and higher income ($35 001 + OR: 4.68, P = .008) were strongly associated with medical app usage. In the wearables model, the same trend was observed with education (bachelor's degree or more OR: 4.45, P = .04), and higher income ($35 001 + OR: 5.49, P = .01). Compared to earlier studies that have reported utilization rates of between 50% and 60%, our finding of much lower utilization in economically disadvantaged populations that are at greater risks for sub-optimal health outcomes gives cause for greater concern. Considering the ongoing proliferation of digital health technological modalities, this further highlights the need to explore and address equity-based barriers to these health tools.

Canine Parvoviral Enteritis Mortality in Shelter Dogs Receiving Conventional Medical Therapy Compared to Integrative Treatment with the Chinese Herbal Medicine, Zhi Li Tang: A Pilot Study

Canine parvoviral (CPV) enteritis is a contagious infection, which particularly affects young dogs, and is associated with high morbidity and mortality rates in the canine population. This study sought to determine whether including the Chinese herbal medicine formula, Zhi Li Tang (ZLT, Red Back Door), in conjunction with conventional therapy can decrease mortality of shelter dogs with CPV enteritis. Nineteen canine subjects from a shelter population diagnosed with CPV enteritis were randomly assigned to a Control Group which received conventional supportive treatment, or an Integrative Treatment Group which received ZLT along with conventional supportive treatment. Study duration was 14 days unless the subject died, or all clinical signs associated with CPV enteritis had resolved. The outcome measurement was a binary response of mortality within the 14-day trial. In the Control Group, there was a 43% (3/7) mortality rate compared to 0% mortality in the Integrative Treatment Group (12/12, all dogs recovered) within the 14-day study period. All subjects that survived in both study groups had resolution of CPV clinical signs by the end of the study. Despite the small sample size in this pilot study, there was a statistically significant (p=0.036) improvement in survival for the dogs receiving Zhi Li Tang in addition to conventional therapy when compared to the dogs receiving conventional therapy only. The results from this study suggest integrating ZLT with conventional treatment of CPV enteritis could reduce mortality from CPV infections in shelter dogs. Further investigation of these findings with larger studies is needed to confirm study results.

Effects of extreme temperature events on deaths and its interaction with air pollution

BackgroundBoth extreme temperature events (ETEs) and air pollution affected human health, and their effects were often not independent. Previous studies have provided limited information on the interactions between ETEs and air pollution. MethodsWe collected data on deaths (non-accidental, cardiovascular, and respiratory) in Zibo City along with daily air pollution and meteorological data from January 2015 to December 2019. Distributed lag non-linear model was used to explore the health effects of ETEs on deaths. Non-parametric binary response model, hierarchical model and joint effect model were used to further explore the interaction between ETEs and air pollution in different seasons. Meanwhile, subgroup analysis by gender and age (≥ 65 years old and < 65 years old) was conducted to identify the vulnerable population. ResultsETEs increased death risk, especially for cardiovascular and respiratory deaths. Heat waves had a stronger impact than cold spells. Cold spells had a longer lag and fluctuating trend. Heat waves had a short-term impact, followed by a decrease. Females and those aged ≥ 65 were more affected, but subgroup differences were not significant. During ETEs and non-ETEs, there were different effects on deaths with per IQR increase in air pollutant concentrations. Joint effect models revealed that there was a significant interaction between ETEs and air pollution on non-accidental deaths. The interaction between PM2.5 and cold spells was antagonistic in the cold season. In the warm season, the health effects of heat waves and high O3 concentration were enhanced. The relative excess risk due to interaction (RERI) of cold spells and PM2.5 in total population was −0.09 (95 % CI: −0.17, −0.01), and 9 % (95 % CI: 1 %, 17 %) of the total effect was attributable to interaction. Subgroup analysis confirmed the interactions in females and those aged ≥ 65. ConclusionsSignificant association observed between ETEs and deaths. Females and ≥ 65 age groups were vulnerable. There were interactions between ETEs and air pollution. The effect of PM2.5 on deaths decreased during cold spells, while the effect of O3 increased during heat waves. In addition to improving air quality, it is necessary to further strengthen the prevention and control of ETEs.

A novel response model and target selection method with applications to marketing

SummaryResponse models used in marketing are not always constructed for later marketing optimisation, which often results in unsatisfactory results in target selection for future marketing activities. To solve this problem, we develop a new binary response model and a new marketing target selection method. The proposed model can predict multiple propensity scores per customer through customer‐specific propensity score distributions, which is not possible with existing response models, filling a gap in the literature. The target selection method can determine the best propensity scores from those predicted by the proposed model and use them to select customers for further marketing activities. Our simulation results and application to real marketing data confirm that the performance of the proposed model in target selection is significantly better than that of the existing models, including some popular machine learning methods, which indicate that our method can be very useful in practice.

Spatial classification in the presence of measurement error

In recent decades, spatial classification has received considerable attention in a wide array of disciplines. In practice, binary response variable is often subject to measurement error, misclassification. To account for the misclassified response in spatial classification, we proposed validation data-based adjustment methods that use interval validation data to rectify misclassified responses. Regression calibration and multiple imputation methods are utilized to correct the misclassified outcomes at the locations where the gold-standard device is not available. Generalized linear mixed model and indicator Kriging are applied for spatial classification at unsampled locations. Simulation studies are performed to compare the proposed methods with naive methods that ignore the misclassification. It was found that the proposed models significantly improve prediction accuracy. Additionally, the proposed models are applied for precipitation detection in South Korea.

Modeling Sequential Dependencies in Progressive Matrices: An Auto-Regressive Item Response Theory (AR-IRT) Approach.

Measurement models traditionally make the assumption that item responses are independent from one another, conditional upon the common factor. They typically explore for violations of this assumption using various methods, but rarely do they account for the possibility that an item predicts the next. Extending the development of auto-regressive models in the context of personality and judgment tests, we propose to extend binary item response models-using, as an example, the 2-parameter logistic (2PL) model-to include auto-regressive sequential dependencies. We motivate such models and illustrate them in the context of a publicly available progressive matrices dataset. We find an auto-regressive lag-1 2PL model to outperform a traditional 2PL model in fit as well as to provide more conservative discrimination parameters and standard errors. We conclude that sequential effects are likely overlooked in the context of cognitive ability testing in general and progressive matrices tests in particular. We discuss extensions, notably models with multiple lag effects and variable lag effects.

2D material platform for overcoming the amplitude–phase tradeoff in ring resonators

Compact and high-speed electro-optic phase modulators play a vital role in various large-scale applications including optical computing, quantum and neural networks, and optical communication links. Conventional electro-refractive phase modulators such as silicon (Si), III-V and graphene on Si suffer from a fundamental tradeoff between device length and optical loss that limits their scaling capabilities. High-finesse ring resonators have been traditionally used as compact intensity modulators, but their use for phase modulation has been limited due to the high insertion loss associated with the phase shift. Here, we show that high-finesse resonators can achieve a strong phase shift with low insertion loss by simultaneous modulation of the real and imaginary parts of the refractive index, to the same extent, i.e., ΔnΔk∼1. To implement this strategy, we demonstrate an active hybrid platform that combines a low-loss SiN ring resonator with 2D materials such as graphene and transition metal dichalcogenide [tungsten disulphide (WSe2)], which induces a strong change in the imaginary and real parts of the index. Our platform consisting of a 25 µm long Gr-Al2O3-WSe2 capacitor embedded on a SiN ring of 50 µm radius (∼8% ring coverage) achieves a continuous phase shift of (0.46±0.05)π radians with an insertion loss (IL) of 3.18±0.20 dB and a transmission modulation (ΔTRing) of 1.72±0.15dB at a probe wavelength (λ p ) of 1646.18 nm. We find that our Gr-Al2O3-WSe2 capacitor exhibits a phase modulation efficiency (Vπ2⋅L) of 0.530±0.016V⋅cm and can support an electro-optic bandwidth of 14.9±0.1GHz. We further show that our platform can achieve a phase shift of π radians with an IL of 5 dB and a minimum ΔT of 0.046 dB. We demonstrate the broadband nature of the binary phase response, by measuring a phase shift of (1.00±0.10)π radians, with an IL of 5.20±0.31dB and a minimal ΔTRing of 0.015±0.006dB for resonances spanning from 1564 to 1650 nm. This SiN–2D hybrid platform provides the design for compact and high-speed reconfigurable circuits with graphene and transition metal dichalcogenide (TMD) monolayers that can enable large-scale photonic systems.

MiMultiCat: A Unified Cloud Platform for the Analysis of Microbiome Data with Multi-Categorical Responses.

The field of the human microbiome is rapidly growing due to the recent advances in high-throughput sequencing technologies. Meanwhile, there have also been many new analytic pipelines, methods and/or tools developed for microbiome data preprocessing and analytics. They are usually focused on microbiome data with continuous (e.g., body mass index) or binary responses (e.g., diseased vs. healthy), yet multi-categorical responses that have more than two categories are also common in reality. In this paper, we introduce a new unified cloud platform, named MiMultiCat, for the analysis of microbiome data with multi-categorical responses. The two main distinguishing features of MiMultiCat are as follows: First, MiMultiCat streamlines a long sequence of microbiome data preprocessing and analytic procedures on user-friendly web interfaces; as such, it is easy to use for many people in various disciplines (e.g., biology, medicine, public health). Second, MiMultiCat performs both association testing and prediction modeling extensively. For association testing, MiMultiCat handles both ecological (e.g., alpha and beta diversity) and taxonomical (e.g., phylum, class, order, family, genus, species) contexts through covariate-adjusted or unadjusted analysis. For prediction modeling, MiMultiCat employs the random forest and gradient boosting algorithms that are well suited to microbiome data while providing nice visual interpretations. We demonstrate its use through the reanalysis of gut microbiome data on obesity with body mass index categories. MiMultiCat is freely available on our web server.

Can feature structure improve model's precision? A novel prediction method using artificial image and image identification.

This study aimed to develop an approach to enhance the model precision by artificial images. Given an epidemiological study designed to predict 1 response using f features with M samples, each feature was converted into a pixel with certain value. Permutated these pixels into F orders, resulting in F distinct artificial image sample sets. Based on the experience of image recognition techniques, appropriate training images results in higher precision model. In the preliminary experiment, a binary response was predicted by 76 features, the sample set included 223 patients and 1776 healthy controls. We randomly selected 10 000 artificial sample sets to train the model. Models' performance (area under the receiver operating characteristic curve values) depicted a bell-shaped distribution. The model construction strategy developed in the research has potential to capture feature order related information and enhance model predictability.

Ultra-high-dimensional feature screening of binary categorical response data based on Jensen-Shannon divergence

&lt;abstract&gt;&lt;p&gt;Currently, most of the ultra-high-dimensional feature screening methods for categorical data are based on the correlation between covariates and response variables, using some statistics as the screening index to screen important covariates. Thus, with the increasing number of data types and model availability limitations, there may be a potential problem with the existence of a class of unimportant covariates that are also highly correlated with the response variable due to their high correlation with the other covariates. To address this issue, in this paper, we establish a model-free feature screening procedure for binary categorical response variables from the perspective of the contribution of features to classification. The idea is to introduce the Jensen-Shannon divergence to measure the difference between the conditional probability distributions of the covariates when the response variables take on different values. The larger the value of the Jensen-Shannon divergence, the stronger the covariate's contribution to the classification of the response variable, and the more important the covariate is. We propose two kinds of model-free ultra-high-dimensional feature screening methods for binary response data. Meanwhile, the methods are suitable for continuous or categorical covariates. When the numbers of covariate categories are the same, the feature screening is based on traditional Jensen-Shannon divergence. When the numbers of covariate categories are different, the Jensen-Shannon divergence is adjusted using the logarithmic factor of the number of categories. We theoretically prove that the proposed methods have sure screening and ranking consistency properties, and through simulations and real data analysis, we demonstrate that, in feature screening, the approaches proposed in this paper have the advantages of effectiveness, stability, and less computing time compared with an existing method.&lt;/p&gt;&lt;/abstract&gt;

Prevalence and Predictors of Adverse Events Associated With Dipeptidyl Peptidase-4 (DPP-4) Inhibitors in Type 2 Diabetic Patients: A Cross-sectional Study

Background: Dipeptidyl peptidase-4 (DPP-4) inhibitors are oral hypoglycemic agents widely prescribed in India despite safety concerns. However, studies focused on their safety profile are scarce, especially in South India. Objective: To evaluate the prevalence and predictors of adverse events (AEs) with DPP-4 inhibitors in patients with type 2 diabetes mellitus (T2DM). Research design and methods: This retrospective cross-sectional study analyzed data from medical records of T2DM patients prescribed DPP-4 inhibitors admitted to the medicine department from 2019 to 2021 at a South Indian tertiary care hospital. The causality of AEs was assessed using the WHO-Uppsala Monitoring Centre (WHO-UMC) criteria and the Naranjo scale, and severity using the Modified Hartwig and Seigel scale. We applied a Generalized model with a binary response and logit-link function to understand the factors that best explain the AE. The best-fit models were chosen based on least Akaike’s information criterion and highest Pseudo R2 and presented the odds ratio (OR) with a 95% confidence interval. The analyses were performed in R software version 4.2.1. Results: Among the 796 patients included in the study, 26% experienced AEs. A total of 212 AEs were observed, and Saxagliptin-associated AEs were the most prevalent (66.6%). Hepatic AEs were predominant (37.7%), followed by gastrointestinal events (16.5%) and electrolyte imbalances (12.3%). Most AEs were possible based on WHO-UMC criteria (78.7%) and the Naranjo scale (86.7%), with 58% being of moderate severity and 42% mild. In the multivariate analysis, aspartate transaminase [OR: 1.013 (0.006–1.020)], alkaline phosphatase [OR: 1.004 (1.001–1.007)] and patients already on DPP-4 inhibitors [OR 1.191(1.012–1.366)] were significant predictors for AEs with DPP-4 inhibitors. Conclusion: The study highlighted a high prevalence of AEs with DPP-4 inhibitors and identified significant predictors of these AEs. These findings underscore the necessity of vigilant monitoring and risk assessment while prescribing DPP-4 inhibitors to the Indian population.

VaR calculation by binary response models

&lt;p&gt;The original Risk-Metrics method is underpinned by the assumption that daily asset returns are conditional Gaussian independently identically distributed (iid) random variables with a mean of zero. In this paper, a new method to calculate Value at Risk (VaR) was suggested to overcome the shortcoming of Risk-Metrics by employing binary response models to compute probability forecasts of the portfolio return by exceeding a grid of candidate quantile values. From those values, the VaR quantile value was selected. The proposed model was called BRV (Binary Response VaR method). Consistent application of BRV to the Dow Jones Industrial Average (INDEXDJX: DJI) and Dow Jones U.S. Marine Transportation Index (DJUSMT) time series proved that it was more accurate than the Risk-Metric system. This method not only worked similar to quantile regression but had the advantage that conventional maximum likelihood methods could be used for parameter estimation and inference. The BRV method was the best performing method for computing the daily VaR at both the 95% and 99% confidence levels over the period 02/01/06–31/12/08. The BRV and the QR (quantile regression) methods performed similarly, but the BRV method had the practical advantage that conventional maximum likelihood (ML) technique could be used for parameter estimation and robust inference.&lt;/p&gt;

A framework for block-wise missing data in multi-omics.

High-throughput technologies have generated vast amounts of omic data. It is a consensus that the integration of diverse omics sources improves predictive models and biomarker discovery. However, managing multiple omics data poses challenges such as data heterogeneity, noise, high-dimensionality and missing data, especially in block-wise patterns. This study addresses the challenges of high dimensionality and block-wise missing data through a regularization and constrained-based approach. The methodology is implemented in the R package bwm for binary and continuous response variables, and applied to breast cancer and exposome multi-omics datasets, achieving strong performance even in scenarios with missing data present in all omics. In binary classification task, our proposed model achieves accuracy in the range of 86% to 92%, and F1 in the range of 68% to 79%. And, in regression task the correlation between true and predicted responses is in the range of 72% to 76%. However, there is a slight decline in performance metrics as the percentage of missing data increases. In scenarios where block-wise missing data affects multiple omics, the model performance actually surpasses that of scenarios where missing data is present in only one omics. One possible explanation for this might be that the other scenarios introduce a greater diversity of observation profiles, leading to a more robust model. Depending on the specific omics being studied, there is greater consistency in feature selection when comparing block-wise missing data scenarios.

Group feature screening for ultrahigh-dimensional data missing at random

&lt;abstract&gt; &lt;p&gt;Statistical inference for missing data is common in data analysis, and there are still widespread cases of missing data in big data. The literature has discussed the practicability of two-stage feature screening with categorical covariates missing at random (IMCSIS). Therefore, we propose group feature screening for ultrahigh-dimensional data with categorical covariates missing at random (GIMCSIS), which can be used to effectively select important features. The proposed method expands the scope of IMCSIS and further improves the performance of classification learning when covariates are missing. Based on the adjusted Pearson chi-square statistics, a two-stage group feature screening method is modeled, and theoretical analysis proves that the proposed method conforms to the sure screening property. In a numerical simulation, GIMCSIS can achieve better finite sample performance under binary and multivariate response variables and multi-classification covariates. The empirical analysis through multiple classification results shows that GIMCSIS is superior to IMCSIS in imbalanced data classification.&lt;/p&gt; &lt;/abstract&gt;