Yunpeng Zhao

Yunpeng Zhao, Fugen Zhou, Bin Guo et al.

Graph neural networks (GNNs) have been successfully applied to early mild cognitive impairment (EMCI) detection, with the usage of elaborately designed features constructed from blood oxygen level-dependent (BOLD) time series. However, few works explored the feasibility of using BOLD signals directly as features. Meanwhile, existing GNN-based methods primarily rely on hand-crafted explicit brain topology as the adjacency matrix, which is not optimal and ignores the implicit topological organization of the brain. In this paper, we propose a spatial temporal graph convolutional network with a novel graph structure self-learning mechanism for EMCI detection. The proposed spatial temporal graph convolution block directly exploits BOLD time series as input features, which provides an interesting view for rsfMRI-based preclinical AD diagnosis. Moreover, our model can adaptively learn the optimal topological structure and refine edge weights with the graph structure self-learning mechanism. Results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database show that our method outperforms state-of-the-art approaches. Biomarkers consistent with previous studies can be extracted from the model, proving the reliable interpretability of our method.

Yunpeng Zhao, Ning Hao, Ji Zhu

Bipartite graphs are ubiquitous across various scientific and engineering fields. Simultaneously grouping the two types of nodes in a bipartite graph via biclustering represents a fundamental challenge in network analysis for such graphs. The latent block model (LBM) is a commonly used model-based tool for biclustering. However, the effectiveness of the LBM is often limited by the influence of row and column sums in the data matrix. To address this limitation, we introduce the degree-corrected latent block model (DC-LBM), which accounts for the varying degrees in row and column clusters, significantly enhancing performance on real-world data sets and simulated data. We develop an efficient variational expectation-maximization algorithm by creating closed-form solutions for parameter estimates in the M steps. Furthermore, we prove the label consistency and the rate of convergence of the variational estimator under the DC-LBM, allowing the expected graph density to approach zero as long as the average expected degrees of rows and columns approach infinity when the size of the graph increases.

Jiao Xu, Junwei Liu, Jiangwei Lao et al.

Recent advances in medical multi-modal models focus on specialized image analysis like dermatology, pathology, or radiology. However, they do not fully capture the complexity of real-world clinical diagnostics, which involve heterogeneous inputs and require ongoing contextual understanding during patient-physician interactions. To bridge this gap, we introduce PulseMind, a new family of multi-modal diagnostic models that integrates a systematically curated dataset, a comprehensive evaluation benchmark, and a tailored training framework. Specifically, we first construct a diagnostic dataset, MediScope, which comprises 98,000 real-world multi-turn consultations and 601,500 medical images, spanning over 10 major clinical departments and more than 200 sub-specialties. Then, to better reflect the requirements of real-world clinical diagnosis, we develop the PulseMind Benchmark, a multi-turn diagnostic consultation benchmark with a four-dimensional evaluation protocol comprising proactiveness, accuracy, usefulness, and language quality. Finally, we design a training framework tailored for multi-modal clinical diagnostics, centered around a core component named Comparison-based Reinforcement Policy Optimization (CRPO). Compared to absolute score rewards, CRPO uses relative preference signals from multi-dimensional com-parisons to provide stable and human-aligned training guidance. Extensive experiments demonstrate that PulseMind achieves competitive performance on both the diagnostic consultation benchmark and public medical benchmarks.

Zhicheng He, Yunpeng Zhao, Junde Wu et al.

Medical image generation is pivotal in applications like data augmentation for low-resource clinical tasks and privacy-preserving data sharing. However, developing a scalable generative backbone for medical imaging requires architectural efficiency, sufficient multi-organ data, and principled evaluation, yet current approaches leave these aspects unresolved. Therefore, we introduce MedVAR, the first autoregressive-based foundation model that adopts the next-scale prediction paradigm to enable fast and scale-up-friendly medical image synthesis. MedVAR generates images in a coarse-to-fine manner and produces structured multi-scale representations suitable for downstream use. To support hierarchical generation, we curate a harmonized dataset of around 440,000 CT and MRI images spanning six anatomical regions. Comprehensive experiments across fidelity, diversity, and scalability show that MedVAR achieves state-of-the-art generative performance and offers a promising architectural direction for future medical generative foundation models.

Yibing Fu, Yunpeng Zhao, Zhitao Zeng et al.

Multi-modal learning integrating medical images and tabular data has significantly advanced clinical decision-making in recent years. Self-Supervised Learning (SSL) has emerged as a powerful paradigm for pretraining these models on large-scale unlabeled image-tabular data, aiming to learn discriminative representations. However, existing SSL methods for image-tabular representation learning are often confined to specific data cohorts, mainly due to their rigid tabular modeling mechanisms when modeling heterogeneous tabular data. This inter-tabular barrier hinders the multi-modal SSL methods from effectively learning transferrable medical knowledge shared across diverse cohorts. In this paper, we propose a novel SSL framework, namely CITab, designed to learn powerful multi-modal feature representations in a cross-tabular manner. We design the tabular modeling mechanism from a semantic-awareness perspective by integrating column headers as semantic cues, which facilitates transferrable knowledge learning and the scalability in utilizing multiple data sources for pretraining. Additionally, we propose a prototype-guided mixture-of-linear layer (P-MoLin) module for tabular feature specialization, empowering the model to effectively handle the heterogeneity of tabular data and explore the underlying medical concepts. We conduct comprehensive evaluations on Alzheimer's disease diagnosis task across three publicly available data cohorts containing 4,461 subjects. Experimental results demonstrate that CITab outperforms state-of-the-art approaches, paving the way for effective and scalable cross-tabular multi-modal learning.

Yunpeng Zhao, Jie Zhang

As synthetic data becomes increasingly popular in machine learning tasks, numerous methods--without formal differential privacy guarantees--use synthetic data for training. These methods often claim, either explicitly or implicitly, to protect the privacy of the original training data. In this work, we explore four different training paradigms: coreset selection, dataset distillation, data-free knowledge distillation, and synthetic data generated from diffusion models. While all these methods utilize synthetic data for training, they lead to vastly different conclusions regarding privacy preservation. We caution that empirical approaches to preserving data privacy require careful and rigorous evaluation; otherwise, they risk providing a false sense of privacy.

Jony Karki, Dongzhou Huang, Yunpeng Zhao

Node popularity is recognized as a key factor in modeling real-world networks, capturing heterogeneity in connectivity across communities. This concept is equally important in bipartite networks, where nodes in different partitions may exhibit varying popularity patterns, motivating models such as the Two-Way Node Popularity Model (TNPM). Existing methods, such as the Two-Stage Divided Cosine (TSDC) algorithm, provide a scalable estimation approach but may have limitations in terms of accuracy or applicability across different types of networks. In this paper, we develop a computationally efficient and theoretically justified variational expectation-maximization (VEM) framework for the TNPM. We establish label consistency for the estimated community assignments produced by the proposed variational estimator in bipartite networks. Through extensive simulation studies, we show that our method achieves superior estimation accuracy across a range of bipartite as well as undirected networks compared to existing algorithms. Finally, we evaluate our method on real-world bipartite and undirected networks, further demonstrating its practical effectiveness and robustness.

Xiang Li, Yunpeng Zhao, Qing Pan et al.

Community detection is the task of clustering objects based on their pairwise relationships. Most of the model-based community detection methods, such as the stochastic block model and its variants, are designed for networks with binary (yes/no) edges. In many practical scenarios, edges often possess continuous weights, spanning positive and negative values, which reflect varying levels of connectivity. To address this challenge, we introduce the heterogeneous block covariance model (HBCM) that defines a community structure within the covariance matrix, where edges have signed and continuous weights. Furthermore, it takes into account the heterogeneity of objects when forming connections with other objects within a community. A novel variational expectation-maximization algorithm is proposed to estimate the group membership. The HBCM provides provable consistent estimates of memberships, and its promising performance is observed in numerical simulations with different setups. The model is applied to a single-cell RNA-seq dataset of a mouse embryo and a stock price dataset. Supplementary materials for this article are available online.

Yunpeng Zhao, Cheng Chen, Qing You Pang et al.

Addressing missing modalities presents a critical challenge in multimodal learning. Current approaches focus on developing models that can handle modality-incomplete inputs during inference, assuming that the full set of modalities are available for all the data during training. This reliance on full-modality data for training limits the use of abundant modality-incomplete samples that are often encountered in practical settings. In this paper, we propose a robust universal model with modality reconstruction and model personalization, which can effectively tackle the missing modality at both training and testing stages. Our method leverages a multimodal masked autoencoder to reconstruct the missing modality and masked patches simultaneously, incorporating an innovative distribution approximation mechanism to fully utilize both modality-complete and modality-incomplete data. The reconstructed modalities then contributes to our designed data-model co-distillation scheme to guide the model learning in the presence of missing modalities. Moreover, we propose a CLIP-driven hyper-network to personalize partial model parameters, enabling the model to adapt to each distinct missing modality scenario. Our method has been extensively validated on two brain tumor segmentation benchmarks. Experimental results demonstrate the promising performance of our method, which consistently exceeds previous state-of-the-art approaches under the all-stage missing modality settings with different missing ratios. Code will be available.

Yefeng Wang, Yunpeng Zhao, Jiang Bian et al.

Objective: The objective of this study is to develop a deep learning pipeline to detect signals on dietary supplement-related adverse events (DS AEs) from Twitter. Material and Methods: We obtained 247,807 tweets ranging from 2012 to 2018 that mentioned both DS and AE. We annotated biomedical entities and relations on 2,000 randomly selected tweets. For the concept extraction task, we compared the performance of traditional word embeddings with SVM, CRF and LSTM-CRF classifiers to BERT models. For the relation extraction task, we compared GloVe vectors with CNN classifiers to BERT models. We chose the best performing models in each task to assemble an end-to-end deep learning pipeline to detect DS AE signals and compared the results to the known DS AEs from a DS knowledge base (i.e., iDISK). Results: In both tasks, the BERT-based models outperformed traditional word embeddings. The best performing concept extraction model is the BioBERT model that can identify supplement, symptom, and body organ entities with F1-scores of 0.8646, 0.8497, and 0.7104, respectively. The best performing relation extraction model is the BERT model that can identify purpose and AE relations with F1-scores of 0.8335 and 0.7538, respectively. The end-to-end pipeline was able to extract DS indication and DS AEs with an F1-score of 0.7459 and 0,7414, respectively. Comparing to the iDISK, we could find both known and novel DS-AEs. Conclusion: We have demonstrated the feasibility of detecting DS AE signals from Twitter with a BioBERT-based deep learning pipeline.

Yunpeng Zhao, Peter Bickel, Charles Weko

Statistical network analysis primarily focuses on inferring the parameters of an observed network. In many applications, especially in the social sciences, the observed data is the groups formed by individual subjects. In these applications, the network is itself a parameter of a statistical model. Zhao and Weko (2019) propose a model-based approach, called the hub model, to infer implicit networks from grouping behavior. The hub model assumes that each member of the group is brought together by a member of the group called the hub. The hub model belongs to the family of Bernoulli mixture models. Identifiability of parameters is a notoriously difficult problem for Bernoulli mixture models. This paper proves identifiability of the hub model parameters and estimation consistency under mild conditions. Furthermore, this paper generalizes the hub model by introducing a model component that allows hubless groups in which individual nodes spontaneously appear independent of any other individual. We refer to this additional component as the null component. The new model bridges the gap between the hub model and the degenerate case of the mixture model -- the Bernoulli product. Identifiability and consistency are also proved for the new model. Numerical studies are provided to demonstrate the theoretical results.

Yunpeng Zhao, Mattia Prosperi, Tianchen Lyu et al.

Social media, especially Twitter, is being increasingly used for research with predictive analytics. In social media studies, natural language processing (NLP) techniques are used in conjunction with expert-based, manual and qualitative analyses. However, social media data are unstructured and must undergo complex manipulation for research use. The manual annotation is the most resource and time-consuming process that multiple expert raters have to reach consensus on every item, but is essential to create gold-standard datasets for training NLP-based machine learning classifiers. To reduce the burden of the manual annotation, yet maintaining its reliability, we devised a crowdsourcing pipeline combined with active learning strategies. We demonstrated its effectiveness through a case study that identifies job loss events from individual tweets. We used Amazon Mechanical Turk platform to recruit annotators from the Internet and designed a number of quality control measures to assure annotation accuracy. We evaluated 4 different active learning strategies (i.e., least confident, entropy, vote entropy, and Kullback-Leibler divergence). The active learning strategies aim at reducing the number of tweets needed to reach a desired performance of automated classification. Results show that crowdsourcing is useful to create high-quality annotations and active learning helps in reducing the number of required tweets, although there was no substantial difference among the strategies tested.

Francois Modave, Yunpeng Zhao, Janice Krieger et al.

Among American women, the rate of breast cancer is only second to lung cancer. An estimated 12.4% women will develop breast cancer over the course of their lifetime. The widespread use of social media across the socio-economic spectrum offers unparalleled ways to facilitate information sharing, in particular as it pertains to health. Social media is also used by many healthcare stakeholders, ranging from government agencies to healthcare industry, to disseminate health information and to engage patients. The purpose of this study is to investigate people's perceptions and attitudes relate to breast cancer, especially those that are related to physical activities, on Twitter. To achieve this, we first identified and collected tweets related to breast cancer; and then used topic modeling and sentiment analysis techniques to understanding discussion themes and quantify Twitter users' perceptions and emotions w.r.t breast cancer to answer 5 research questions.

Yunpeng Zhao, Qing Pan, Chengan Du

When searching for gene pathways leading to specific disease outcomes, additional information on gene characteristics is often available that may facilitate to differentiate genes related to the disease from irrelevant background when connections involving both types of genes are observed and their relationships to the disease are unknown. We propose method to single out irrelevant background genes with the help of auxiliary information through a logistic regression, and cluster relevant genes into cohesive groups using the adjacency matrix. Expectation-maximization algorithm is modified to maximize a joint pseudo-likelihood assuming latent indicators for relevance to the disease and latent group memberships as well as Poisson or multinomial distributed link numbers within and between groups. A robust version allowing arbitrary linkage patterns within the background is further derived. Asymptotic consistency of label assignments under the stochastic blockmodel is proven. Superior performance and robustness in finite samples are observed in simulation studies. The proposed robust method identifies previously missed gene sets underlying autism related neurological diseases using diverse data sources including de novo mutations, gene expressions and protein-protein interactions.

Yunpeng Zhao

Real-world networks usually have community structure, that is, nodes are grouped into densely connected communities. Community detection is one of the most popular and best-studied research topics in network science and has attracted attention in many different fields, including computer science, statistics, social sciences, among others. Numerous approaches for community detection have been proposed in literature, from ad-hoc algorithms to systematic model-based approaches. The large number of available methods leads to a fundamental question: whether a certain method can provide consistent estimates of community labels. The stochastic blockmodel (SBM) and its variants provide a convenient framework for the study of such problems. This article is a survey on the recent theoretical advances of community detection. The authors review a number of community detection methods and their theoretical properties, including graph cut methods, profile likelihoods, the pseudo-likelihood method, the variational method, belief propagation, spectral clustering, and semidefinite relaxations of the SBM. The authors also briefly discuss other research topics in community detection such as robust community detection, community detection with nodal covariates and model selection, as well as suggest a few possible directions for future research.

Yunpeng Zhao

In social network analysis, the observed data is usually some social behavior, such as the formation of groups, rather than an explicit network structure. Zhao and Weko (2017) propose a model-based approach called the hub model to infer implicit networks from grouped observations. The hub model assumes independence between groups, which sometimes is not valid in practice. In this article, we generalize the idea of the hub model into the case of grouped observations with temporal dependence. As in the hub model, we assume that the group at each time point is gathered by one leader. Unlike in the hub model, the group leaders are not sampled independently but follow a Markov chain, and other members in adjacent groups can also be correlated. An expectation-maximization (EM) algorithm is developed for this model and a polynomial-time algorithm is proposed for the E-step. The performance of the new model is evaluated under different simulation settings. We apply this model to a data set of the Kibale Chimpanzee Project.

Chengan Du, Yunpeng Zhao, Feng Wang

Graph-based semi-supervised learning is one of the most popular methods in machine learning. Some of its theoretical properties such as bounds for the generalization error and the convergence of the graph Laplacian regularizer have been studied in computer science and statistics literatures. However, a fundamental statistical property, the consistency of the estimator from this method has not been proved. In this article, we study the consistency problem under a non-parametric framework. We prove the consistency of graph-based learning in the case that the estimated scores are enforced to be equal to the observed responses for the labeled data. The sample sizes of both labeled and unlabeled data are allowed to grow in this result. When the estimated scores are not required to be equal to the observed responses, a tuning parameter is used to balance the loss function and the graph Laplacian regularizer. We give a counterexample demonstrating that the estimator for this case can be inconsistent. The theoretical findings are supported by numerical studies.

Yunpeng Zhao, Elizaveta Levina, Ji Zhu

Link prediction is one of the fundamental problems in network analysis. In many applications, notably in genetics, a partially observed network may not contain any negative examples of absent edges, which creates a difficulty for many existing supervised learning approaches. We develop a new method which treats the observed network as a sample of the true network with different sampling rates for positive and negative examples. We obtain a relative ranking of potential links by their probabilities, utilizing information on node covariates as well as on network topology. Empirically, the method performs well under many settings, including when the observed network is sparse. We apply the method to a protein-protein interaction network and a school friendship network.