Distribution Divergence Research Articles

Target informed client recruitment for efficient federated learning in healthcare

BackgroundModern machine learning and deep learning methods have been widely incorporated in decision making processes in healthcare in the form of decision support mechanisms. In healthcare, data are abundant but typically not centrally available and, therefore, require some form of aggregation to facilitate training procedures. Aggregating sensitive data poses a significant privacy risk, which is why, both in Europe and the United States, legal frameworks regulate the treatment of such data. Whilst these measures protect the individual behind the data, they pose a significant challenge that results in extensive legal administration related to data sharing efforts. Federated learning (FL) offers a way to mitigate these challenges by allowing to learn models in distributed fashion, eliminating the need to aggregate data for the purpose of training. However, FL comes with a new set of challenges related to communication overhead, client selection and efficiency of the FL training procedure, among others.MethodsIn this work, we extend on a previously proposed client recruitment approach by incorporating knowledge on the local hardware such that it becomes possible to recruit a subset of clients for the federation based on the construct of client-level representativeness, which is expressed in terms of the local target distribution divergence, sample size, and the underlying hardware.ResultsWe show that, for prominent, medical regression and classification tasks, the recruitment approach yields results that are on par, or better, compared to the central and federated approaches. The proposed approach requires a mere fraction of the data for training and reduces the training time by a factor of 3-4. In addition, we show that excluded clients can still significantly benefit from the resulting federated model through local fine-tuning.ConclusionsBy expressing the representativeness of clients in function of the deviation in the local target distribution, the sample size and efficiency of the underlying hardware, we are able to define a recruitment approach that yields a subset of clients for the federation resulting in significantly reduced training time, without harming predictive performance, whilst improving the privacy preserving characteristics compared to the standard FL and central approaches.

Triplet Adaptation Framework for Robust Semi-Supervised Learning.

Semi-supervised learning (SSL) suffers from severe performance degradation when labeled and unlabeled data come from inconsistent and imbalanced distribution. Nonetheless, there is a lack of theoretical guidance regarding a remedy for this issue. To bridge the gap between theoretical insights and practical solutions, we embark to an analysis of generalization bound of classic SSL algorithms. This analysis reveals that distribution inconsistency between unlabeled and labeled data can cause a significant generalization error bound. Motivated by this theoretical insight, we present a Triplet Adaptation Framework (TAF) to reduce the distribution divergence and improve the generalization of SSL models. TAF comprises three adapters: Balanced Residual Adapter, aiming to map the class distribution of labeled and unlabeled data to a uniform distribution for reducing class distribution divergence; Representation Adapter, aiming to map the representation distribution of unlabeled data to labeled one for reducing representation distribution divergence; and Pseudo-Label Adapter, aiming to align the predicted pseudo-labels with the class distribution of unlabeled data, thereby preventing erroneous pseudo-labels from exacerbating representation divergence. These three adapters collaborate synergistically to reduce the generalization bound, ultimately achieving a more robust and generalizable SSL model. Extensive experiments across various robust SSL scenarios validate the efficacy of our method.

SEA++: Multi-Graph-Based Higher-Order Sensor Alignment for Multivariate Time-Series Unsupervised Domain Adaptation.

Unsupervised Domain Adaptation (UDA) methods have been successful in reducing label dependency by minimizing the domain discrepancy between labeled source domains and unlabeled target domains. However, these methods face challenges when dealing with Multivariate Time-Series (MTS) data. MTS data typically originates from multiple sensors, each with its unique distribution. This property poses difficulties in adapting existing UDA techniques, which mainly focus on aligning global features while overlooking the distribution discrepancies at the sensor level, thus limiting their effectiveness for MTS data. To address this issue, a practical domain adaptation scenario is formulated as Multivariate Time-Series Unsupervised Domain Adaptation (MTS-UDA). In this paper, we propose SEnsor Alignment (SEA) for MTS-UDA, aiming to address domain discrepancy at both local and global sensor levels. At the local sensor level, we design endo-feature alignment, which aligns sensor features and their correlations across domains. To reduce domain discrepancy at the global sensor level, we design exo-feature alignment that enforces restrictions on global sensor features. We further extend SEA to SEA++ by enhancing the endo-feature alignment. Particularly, we incorporate multi-graph-based higher-order alignment for both sensor features and their correlations. Extensive empirical results have demonstrated the state-of-the-art performance of our SEA and SEA++ on six public MTS datasets for MTS-UDA.

Digital twin-assisted fault diagnosis framework for rolling bearings under imbalanced data

The application of deep learning-based fault diagnosis methods is constrained by the imbalanced data. Recently, many studies have suggested integrating dynamic model responses into the training process to address data imbalances. However, significant distribution discrepancies exist between dynamic model responses and real measured data, resulting in suboptimal performance. To address this challenge, this research proposes a digital twin-assisted framework for rolling bearings fault diagnosis under imbalanced data, which minimizes the distribution discrepancies between dynamic model responses and real measured data through information and feature transfer. Firstly, a Digital Twin-assisted Data Fusion Strategy (DTDFS) is proposed to facilitate information transfer from physical entities to dynamic models, generating digital twin data for data augmentation. Subsequently, a Frequency Filter Subdomain Adaptation Network (FFSAN) is proposed to achieve feature transfer between twin data and measured data. Finally, experimental results and engineering applications demonstrate that the proposed framework significantly outperforms existing imbalanced fault diagnosis methods, which is crucial to the application of deep learning-based fault diagnosis in industrial settings.

FTAN: Feature Transform and Alignment Network for cross-domain specific emitter identification

Conventional deep learning-based specific emitter identification (SEI) methods are consistently constrained to domain-invariant assumption, leading to a decrease in recognition accuracy when the feature domain changes. To tackle this issue, we propose a novel unsupervised domain adaptation (UDA) framework named feature transform and alignment network (FTAN) for cross-domain SEI. In FTAN, we first apply a weight-shared network to extract the initial features of signals from all domains. Then, we introduce domain-specific modules to individually learn domain-invariant features, which can minimize the distribution discrepancies of source and target signals. Finally, the aligned domain-invariant features are utilized for identification. We evaluate the performance of FTAN on the various signal datasets. The experimental results demonstrate that FTAN significantly mitigates identification performance degradation in cross-domain scenarios and outperforms other state-of-the-art methods.

Audio-Based Engine Fault Diagnosis with Wavelet, Markov Blanket, ROCKET, and Optimized Machine Learning Classifiers.

Engine fault diagnosis is a critical task in automotive aftermarket management. Developing appropriate fault-labeled datasets can be challenging due to nonlinearity variations and divergence in feature distribution among different engine kinds or operating scenarios. To solve this task, this study experimentally measures audio emission signals from compression ignition engines in different vehicles, simulating injector failures, intake hose failures, and absence of failures. Based on these faults, a hybrid approach is applied to classify different conditions that help the planning and decision-making of the automobile industry. The proposed hybrid approach combines the wavelet packet transform (WPT), Markov blanket feature selection, random convolutional kernel transform (ROCKET), tree-structured Parzen estimator (TPE) for hyperparameters tuning, and ten machine learning (ML) classifiers, such as ridge regression, quadratic discriminant analysis (QDA), naive Bayes, k-nearest neighbors (k-NN), support vector machine (SVM), multilayer perceptron (MLP), random forest (RF), extra trees (ET), gradient boosting machine (GBM), and LightGBM. The audio data are broken down into sub-time series with various frequencies and resolutions using the WPT. These data are subsequently utilized as input for obtaining an informative feature subset using a Markov blanket-based selection method. This feature subset is then fed into the ROCKET method, which is paired with ML classifiers, and tuned using Optuna using the TPE approach. The generalization performance applying the proposed hybrid approach outperforms other standard ML classifiers.

Lifting wavelet-informed hierarchical domain adaptation network: An interpretable digital twin-driven gearbox fault diagnosis method

Digital twin (DT) has served as a dependable technology for supplementing reliable simulated fault data in gearbox fault diagnosis. However, the vast data distribution discrepancy and insufficient interpretability still significantly limit the industrial application of DT-driven fault diagnosis methods. To solve these problems, a lifting wavelet-informed hierarchical domain adaptation network (LHDAN) is proposed for transferring the diagnostic knowledge between the physical gearbox and DT model. LHDAN improves the interpretability of diagnostic knowledge transfer in terms of parameter initialization, physical constraints on the training process, and feature distribution adaptation. Specifically, LHDAN utilizes a lifting wavelet-informed convolutional neural network (LW-Conv) to mimic the cascade structure of lifting wavelet decomposition, in which the fully learnable prediction and update operators are initialized with existing wavelet bases and further constrained with high-pass and low-pass filters in the training process. Furthermore, a kurtosis-guided attention mechanism is proposed to fuse hierarchical features with diverse transferabilities flexibly. Finally, the fused hierarchical features of the actual gearbox and DT model are explicitly aligned to eliminate the feature distribution discrepancies. A high-fidelity DT model is established based on an industrial gearbox fault test bench. Compared to several state-of-the-art models, LHDAN demonstrates superior interpretability and diagnostic performance.

Select, Purify, and Exchange: A Multisource Unsupervised Domain Adaptation Method for Building Extraction.

Accurately extracting buildings from aerial images has essential research significance for timely understanding human intervention on the land. The distribution discrepancies between diversified unlabeled remote sensing images (changes in imaging sensor, location, and environment) and labeled historical images significantly degrade the generalization performance of deep learning algorithms. Unsupervised domain adaptation (UDA) algorithms have recently been proposed to eliminate the distribution discrepancies without re-annotating training data for new domains. Nevertheless, due to the limited information provided by a single-source domain, single-source UDA (SSUDA) is not an optimal choice when multitemporal and multiregion remote sensing images are available. We propose a multisource UDA (MSUDA) framework SPENet for building extraction, aiming at selecting, purifying, and exchanging information from multisource domains to better adapt the model to the target domain. Specifically, the framework effectively utilizes richer knowledge by extracting target-relevant information from multiple-source domains, purifying target domain information with low-level features of buildings, and exchanging target domain information in an interactive learning manner. Extensive experiments and ablation studies constructed on 12 city datasets prove the effectiveness of our method against existing state-of-the-art methods, e.g., our method achieves 59.1% intersection over union (IoU) on Austin and Kitsap → Potsdam, which surpasses the target domain supervised method by 2.2% . The code is available at https://github.com/QZangXDU/SPENet.

Plasma Fibrinogen Level’s: A Comparative Study Between Newly Diagnosed Diabetics and Healthy Individuals

The study examines the plasma fibrinogen levels in patients with newly diagnosed Type 2 Diabetes Mellitus (T2DM) compared to healthy individuals at Aarupadai Veedu Medical College in Puducherry. Plasma fibrinogen, an essential protein involved in the process of blood clotting, also functions as a biomarker for cardiovascular and inflammatory illnesses. Employing a comparative cross-sectional approach, we assessed the levels of fibrinogen in a group of 57 patients with newly diagnosed type 2 diabetes mellitus (T2DM) and a group of 57 healthy individuals who were in non-diabetic. The findings indicate a substantial increase in fibrinogen levels among individuals with newly diagnosed type 2 diabetes mellitus (T2DM) at 384.66 mg/dL, in contrast to the control group at 287.24 mg/dL, with a p-value of less than 0.0001. The age distribution discrepancies were found to be statistically significant (p=0.007), indicating that age may be a confounding factor. The gender distribution was balanced, suggesting the absence of any gender-related prejudice. The results of our study emphasise the increased levels of fibrinogen in patients with newly diagnosed type 2 diabetes mellitus (T2DM) and emphasise the necessity for additional investigation into its clinical significance and potential as a target for therapy.

Relationship between Residential Patterns and Socioeconomic Statuses Based on Multi-Source Spatial Data: A Case Study of Nanjing, China

The relationship between residential patterns and socioeconomic statuses highlights the complex interactions between the economic regime, welfare system, and neighborhood effects, which are crucial in urban inequality studies. With the diversification of the housing demand and supply system, the traditional analysis conducted separately from the ethnic or spatial segregation perspective fails to capture the rising inequalities and changing socio-spatial context. Taking Nanjing as an example, based on a multi-source database including the housing price, residential environmental quality, surrounding support facilities, and mobile phone user portrait data, this paper proposed a modified method for discovering the coupling relationship between residential patterns and socioeconomic statuses. It is found that socioeconomic status contributes to residential spatial aggregation and that the relationship between social and spatial dimensions of residential differentiation is tightly coupled and related. The lower socioeconomic strata were displaced to the periphery and the older urban core, while affluent inhabitants were more likely to settle voluntarily in segregated enclaves to isolate themselves from the general population through more flexible housing options. The heterogeneity of the urban socioeconomic dimension is primarily affected by consumption and occupational status, while housing prices mainly determine the divergence of spatial distribution.

Cascaded Cross-modal Alignment for Visible-Infrared Person Re-Identification

Visible-Infrared Person Re-Identification faces significant challenges due to cross-modal and intra-modal variations. Although existing methods explore semantic alignment from various angles, severe distribution shifts in heterogeneous data limit the effectiveness of single-level alignment approaches. To address this issue, we propose a Cascaded Cross-modal Alignment (CCA) framework that gradually eliminates distribution discrepancies and aligns semantic features from three complementary perspectives in a cascaded manner. First, at the input-level, we propose a Channel-Spatial Recombination (CSR) strategy that strategically reorganizes and preserves crucial details from channel and spatial dimensions to diminish visual discrepancies between modalities, thereby narrowing the modality gap in input images. Second, at the frequency-level, we introduce a Low Frequency Masking (LFM) module to emphasize global details that CSR might overlook by randomly masking low-frequency information, thus driving comprehensive alignment of identity semantics. Third, at the part-level, we design a Prototype-based Semantic Refinement (PSR) module to refine fine-grained features and mitigate the impact of irrelevant areas in LFM. It accurately aligns body parts and enhances semantic consistency guided by global discriminative clues from LFM and flipped views with pose variations. Comprehensive experimental results on the SYSU-MM01 and RegDB datasets demonstrate the superiority of our proposed CCA.

A novel cross-receptive field fusion cascade network with adaptive mask update for transfer health state diagnosis of manipulators

Manipulators, particularly planar parallel manipulators, are widely employed in high-end precision equipment to conduct precise positioning and operation tasks due to their advantages of high stiffness, high precision, and high load. Moreover, they are also frequently exposed to changeable working circumstances, which significantly cause inconsistent health state data distribution. Although transfer learning can successfully offset the above distribution discrepancies, it remains unclear how to identify and quantify the source domain knowledge’s contribution to the transfer process. To overcome these challenges, a novel transfer health state diagnosis framework, named cross-receptive field fusion cascade network with adaptive mask update (CFFCN-AMU), is developed and employed for manipulators. Specifically, a unique cross-receptive field fusion cascade module (CFFCM), in which the receptive field self-evaluator and channel attention mechanism are jointly designed, is constructed initially to achieve adaptive extraction and fusion of cascaded features. Subsequently, in the target domain fine-tuning stage, an adaptive mask update (AMU) strategy is implemented to evaluate the contribution of source domain knowledge and selectively guide the parameter updating process. Finally, some mechanistic model-driven cross-working condition transfer scenarios are investigated. Multiple sets of excellent transfer diagnosis results fully illustrate the transferability and superiority of the constructed CFFCN-AMU model.

Multisource partial domain adaptation for bearing fault diagnosis

Abstract Domain adaptation has been widely used in fault diagnosis and dealt with data distribution discrepancies, but the labels in source and target domains are usually assumed to be identical. The complexity of the working conditions, in reality, leads to the fact that the labels in target domains are often a subset of the labels in source domains. This special case is called the partial domain problem. However, most of the existing proposed methods for solving partial domain problems are limited to single-source-domain scenarios and fail to effectively integrate multisource knowledge. Hence, this study proposes a new approach of multisource domain obfuscation-subdomain alignment (MSDO-SA) for partial domain adaptation fault diagnosis in multisource domains. Through domain obfuscation, the multisource domains are converted into a single source domain. The subdomain alignment aims at improving the generalization ability and relevance of the model, and effectively alleviates the domain shift problem. Finally, multiple partial domain fault diagnosis tasks using the CWRU dataset validate the effectiveness, robustness, and superiority of the proposed method.

Unsupervised health indicator construction by a new Gaussian-student’s t-distribution mixture model and its application

The equipment’s remaining useful life (RUL) must be accurately estimated to guarantee its reliable operation. As a crucial part of data-driven RUL prediction, the health indicator (HI) construction method employing the distribution discrepancies can represent the variation trend of health conditions. However, the existing Gaussian mixture model based HI construction method cannot accurately estimate the long-tail distribution characteristics in some degradation data. Moreover, it cannot comprehensively mine the distribution characteristics of degradation data by leveraging different types of distributions. A novel Gaussian-student’s t-distribution mixture model (GSMM) that simultaneously considers Gaussian distribution and student’s t-distribution is developed in this work to estimate the distributions of normal and degradation data. Next, the distribution contact ratio metric (DCRM) is applied to measure the discrepancies between the baseline distribution of normal data and the distributions of test data at different moments. The bearing HI can be constructed with the acquired DCRMs. Finally, the effectiveness and merit of the developed HI construction approach are validated by two bearing life-cycle datasets. The experimental results illustrate that the GSMM-based HI performs better than other classical and state-of-the-art HIs. Additionally, the constructed HI is more suitable for bearing RUL prediction.

Simulation data-driven adaptive frequency filtering focal network for rolling bearing fault diagnosis

The scarcity of well-labeled samples severely limits the application of deep learning-based fault diagnosis methods. To address this issue, this paper proposes a novel domain-adaptive intelligent diagnostic method, termed a simulation data-driven adaptive frequency filtering focal network, which transfers knowledge from dynamic simulation data to unlabeled measured data. First, a dynamic simulation model is established to simulate the coupled behavior of rolling bearings and generate substantial labeled simulation data. Subsequently, an adaptive frequency filter is introduced to suppress random interference components in the frequency domain, complementing conventional time-domain feature extraction methods, thereby reducing the distribution discrepancy between simulation and measured data. Finally, a domain-aware weighted focusing strategy is proposed to adjust sample weights based on predicted domain labels, helping the model focus on difficult samples with significant distribution discrepancies and reduce misclassification. Experimental results demonstrate that the proposed method effectively integrates simulation and measured data, achieving high-accuracy fault diagnosis of rolling bearings in unlabeled measured data. The proposed approach offers a promising fault diagnosis solution in scenarios where obtaining labeled samples is challenging or impractical.

Robust Remote Sensing Scene Interpretation Based on Unsupervised Domain Adaptation

Deep learning models excel in interpreting the exponentially growing amounts of remote sensing data; however, they are susceptible to deception and spoofing by adversarial samples, posing catastrophic threats. The existing methods to combat adversarial samples have limited performance in robustness and efficiency, particularly in complex remote sensing scenarios. To tackle these challenges, an unsupervised domain adaptation algorithm is proposed for the accurate identification of clean images and adversarial samples by exploring a robust generative adversarial classification network that can harmonize the features between clean images and adversarial samples to minimize distribution discrepancies. Furthermore, linear polynomial loss as a replacement for cross-entropy loss is integrated to guide robust representation learning. Additionally, we leverage the fast gradient sign method (FGSM) and projected gradient descent (PGD) algorithms to generate adversarial samples with varying perturbation amplitudes to assess model robustness. A series of experiments was performed on the RSSCN7 dataset and SIRI-WHU dataset. Our experimental results illustrate that the proposed algorithm performs exceptionally well in classifying clean images while demonstrating robustness against adversarial perturbations.

An electronic nose drift compensation algorithm based on semi-supervised adversarial domain adaptive convolutional neural network

Sensor drift is a significant challenge leading to performance degradation in electronic nose(E-nose) systems. Effectively addressing sensor drift represents the most daunting problem in E-nose technology. This work proposes a domain adaptive approach called Semi-Supervised Adversarial Domain Adaptive Convolutional Neural Network (SAD-CNN) to tackle the long-term drift in E-nose and device displacement. SAD-CNN leverages adversarial learning to minimize distribution disparities between the source and target domains. Unlike traditional methods employing projection matrices, SAD-CNN utilizes one-dimensional convolutional neural networks as feature extractors, circumventing the complexities associated with parameter adjustments and matrix calculations in the projection process. During the training process, utilizing pseudo-labels generated by the semi-supervised self-training method to train the model, thereby reducing the labeling costs. Additionally, confidence threshold screening is introduced during the self-training phase to minimize erroneous pseudo-labels. Furthermore, the regenerative Hilbert space’s Maximum Mean Difference combined with Minimum Variance is introduced as a domain constraint function to mitigate distribution discrepancies between domains and enhance feature discriminability across domains. The experimental results demonstrate that the SAD-CNN method outperforms others. Within the long-term drift dataset, the classification accuracies for different scenarios are 78.01 % and 82.53 %, respectively. Meanwhile, the instrument change dataset yields a classification accuracy of 96.45 %.

Holocene Climate Change Promoted Allopatric Divergence and Disjunct Geographic Distribution in a Bee Orchid Species

ABSTRACTAimSpecies with disjunct geographic distributions provide natural opportunities to investigate incipient or recent allopatric divergence. The combination of both genetic and ecological data may be fruitful to decipher the causes of such patterns: (i) actual vicariance, (ii) successful colonisation from one source to a new range (dispersal, biological introduction) or (iii) parallel convergent evolution.LocationSouthern France and Northern Spain.TaxonThe bee orchid Ophrys aveyronensis (and its two recognised subspecies O. a. subsp. aveyronensis and O. a. subsp. vitorica) displays a disjunct geographic distribution with two subranges separated by 600 km on both sides of the Pyrenees mountain range.MethodsAs allopatric divergence is often complex to document in the wild, we used a combination of population genomics and ecological niche modelling (ENM) to investigate the causes of this intriguing biogeographic pattern.ResultsThe population genomic data demonstrate that all the studied populations exhibit similar patterns of genetic diversity and dramatic decrease in effective size compared with the ancestral population. Significant genetic differentiation and reciprocal monophyly exist between populations of the two subranges of O. aveyronensis, despite a very recent divergence time as young as ca. 1500 generations ago. Moreover, paleo‐ENM analyses support that the disjunct geographic distribution of O. aveyronensis is consistent with a range split of a broad ancestral range, contraction and distinct longitudinal and latitudinal shifts in response to climate warming during the Holocene.Main ConclusionThe congruence of the results obtained from both population genomics and ENM approaches documents how very recent continental allopatric divergence initiated speciation in this system. O. aveyronensis provides a promising opportunity to study the onset of reproductive isolation and parallel evolution following an initial stage of geographic separation in a group with high diversification rate.

Discrepancy-based diffusion models for lesion detection in brain MRI

Diffusion probabilistic models (DPMs) have exhibited significant effectiveness in computer vision tasks, particularly in image generation. However, their notable performance heavily relies on labelled datasets, which limits their application in medical images due to the associated high-cost annotations. Current DPM-related methods for lesion detection in medical imaging, which can be categorized into two distinct approaches, primarily rely on image-level annotations. The first approach, based on anomaly detection, involves learning reference healthy brain representations and identifying anomalies based on the difference in inference results. In contrast, the second approach, resembling a segmentation task, employs only the original brain multi-modalities as prior information for generating pixel-level annotations. In this paper, our proposed model – discrepancy distribution medical diffusion (DDMD) – for lesion detection in brain MRI introduces a novel framework by incorporating distinctive discrepancy features, deviating from the conventional direct reliance on image-level annotations or the original brain modalities. In our method, the inconsistency in image-level annotations is translated into distribution discrepancies among heterogeneous samples while preserving information within homogeneous samples. This property retains pixel-wise uncertainty and facilitates an implicit ensemble of segmentation, ultimately enhancing the overall detection performance. Thorough experiments conducted on the BRATS2020 benchmark dataset containing multimodal MRI scans for brain tumour detection demonstrate the great performance of our approach in comparison to state-of-the-art methods.

Inter-participant transfer learning with attention based domain adversarial training for P300 detection

A Brain-computer interface (BCI) system establishes a novel communication channel between the human brain and a computer. Most event related potential-based BCI applications make use of decoding models, which requires training. This training process is often time-consuming and inconvenient for new users. In recent years, deep learning models, especially participant-independent models, have garnered significant attention in the domain of ERP classification. However, individual differences in EEG signals hamper model generalization, as the ERP component and other aspects of the EEG signal vary across participants, even when they are exposed to the same stimuli. This paper proposes a novel One-source domain transfer learning method based Attention Domain Adversarial Neural Network (OADANN) to mitigate data distribution discrepancies for cross-participant classification tasks. We train and validate our proposed model on both a publicly available OpenBMI dataset and a Self-collected dataset, employing a leave one participant out cross validation scheme. Experimental results demonstrate that the proposed OADANN method achieves the highest and most robust classification performance and exhibits significant improvements when compared to baseline methods (CNN, EEGNet, ShallowNet, DeepCovNet) and domain generalization methods (ERM, Mixup, and Groupdro). These findings underscore the efficacy of our proposed method.