Single-cell RNA sequencing (scRNA-seq) enables high-resolution characterization of cellular heterogeneity, but its rich, complementary structure across cells and genes remains underexploited, especially in the presence of technical noise and sparsity. Effectively leveraging this multi-scale structure is essentially an information fusion problem that requires integrating heterogeneous graph-based views of cells and genes into robust low-dimensional representations. In this paper, we introduce GatorSC, a unified representation learning framework that models scRNA-seq data through multi-scale cell and gene graphs and fuses them with a mixture-of-experts architecture. GatorSC constructs a global cell-cell graph, a global gene-gene graph, and a local gene-gene graph derived from neighborhood-specific subgraphs, and learns graph neural network embeddings that are adaptively fused by a gating network. To learn noise-robust and structure-preserving embeddings without labels, we couple graph reconstruction and graph contrastive learning in a unified self-supervised objective applied to both cell- and gene-level graphs. We evaluate GatorSC on 19 publicly available scRNA-seq datasets covering diverse tissues, species, and sequencing platforms. Experiments showed that GatorSC consistently outperforms state-of-the-art deep generative, graph-based, and contrastive methods for cell clustering, gene expression imputation, and cell-type annotation. The learned embeddings are used for accurate trajectory inference, recovery of canonical marker gene programs, and cell-type-specific pathway signatures in an Alzheimer's disease single-nucleus dataset. GatorSC provides a flexible foundation for comprehensive single-cell transcriptomic analysis and can be readily extended to multi-omic and spatial modalities.

Single-cell RNA sequencing (scRNA-seq) offers unprecedented insights into transcriptome-wide gene expression at the single-cell level. Cell clustering has been long established in the analysis of scRNA-seq data to identify the groups of cells with similar expression profiles. However, cell clustering is technically challenging, as raw scRNA-seq data have various analytical issues, including high dimensionality and dropout values. Existing research has developed deep learning models, such as graph machine learning models and contrastive learning-based models, for cell clustering using scRNA-seq data and has summarized the unsupervised learning of cell clustering into a human-interpretable format. While advances in cell clustering have been profound, we are no closer to finding a simple yet effective framework for learning high-quality representations necessary for robust clustering. In this study, we propose scSimGCL, a novel framework based on the graph contrastive learning paradigm for self-supervised pretraining of graph neural networks. This framework facilitates the generation of high-quality representations crucial for cell clustering. Our scSimGCL incorporates cell-cell graph structure and contrastive learning to enhance the performance of cell clustering. Extensive experimental results on simulated and real scRNA-seq datasets suggest the superiority of the proposed scSimGCL. Moreover, clustering assignment analysis confirms the general applicability of scSimGCL, including state-of-the-art clustering algorithms. Further, ablation study and hyperparameter analysis suggest the efficacy of our network architecture with the robustness of decisions in the self-supervised learning setting. The proposed scSimGCL can serve as a robust framework for practitioners developing tools for cell clustering. The source code of scSimGCL is publicly available at https://github.com/zhangzh1328/scSimGCL.

Health risk refers to the probability of the occurrence of a specific health outcome for a specific patient. Interest in health risk prediction has been increasing, especially with the availability of a large amount of electronic health records (EHR). An EHR contains multivariate time series data that records meaningful information associated with a chronological set of clinical events for each patient. Recurrent neural networks (RNN) and hidden Markov models (HMM) have been widely used as generative models of time series data. RNN-based models have strong prediction performance but lack transparency. HMMs have a simple functional form and the ability to provide an intuitive probabilistic interpretation, but their state dynamics are 'memoryless', making it difficult to thoroughly take into account the irregularity in patients' health trajectory. This paper proposes a novel deep Markov network for health risk prediction. The method integrates two modules, a GRU (Gated Recurrent Unit) with attention mechanism and a Neural HMM, into a single network. The GRU generates the inputs required for health risk predictions and uses an attention mechanism to create memorable state dynamics for the Neural HMM. The Neural HMM then provides interpretable structured representations through training. Mixture Density Networks are incorporated in the Neural HMM, which contribute to the modeling of complex patterns found in the transition process. Furthermore, an inference network is designed to embed hidden state representations of GRU and Neural HMM into the same space. The inference network enables the two types of representations to learn from each other during the decoding process of Neural HMM, thereby improving the quality of interpretable structured representations. Experimental results on MIMIC-III and eICU datasets demonstrate that our method can outperform state-of-the-art methods and provide transparency of the model decisions.

Electronic health records (EHRs) have an inherently high degree of irregularity, including many missing values and varying time intervals, due to variations in patient conditions and treatment needs. This makes successful health risk prediction challenging. EHRs contain longitudinal patient data that records meaningful information associated with a chronological set of clinical observations for each patient. Existing methods focus on modeling variable correlations in patient data with deep neural networks to impute missing values and feed complete data matrices into machine learning models to perform downstream healthcare prediction tasks. However, not enough attention was given to the reliability of the imputed values by these methods. Further, it is likely that the pattern of missing data in EHR contains important information affecting relationships among variables, including time intervals. We propose a novel deep imputation-prediction network to simultaneously perform imputation and prediction tasks with EHR. Our method has the advantages of being able to: 1) learn from the longitudinal patient data in both forward and backward directions, 2) generate both the predicted and imputed values and enhance the reliability of imputed values, and 3) incorporate three common decay functions to capture the variation pattern of input variables in time and adaptively enhances the temporal representation of each pattern with adjustable weights. As well, our method is able to examine the association between input variables to identify critical indicative variables regardless of how long ago the associated event happened. Experimental results on MIMIC-III and eICU datasets demonstrate the effectiveness and superiority of our method for both imputation and prediction, as well as transparency and interpretability, compared to existing state-of-the-art methods.

Predicting the health risks of patients based on electronic health records (EHRs) has recently attracted considerable research interest. Health risk refers to the probability of the occurrence of a specific health outcome for a specific patient. The predicted risks of a specific health outcome can be used to support decisions by healthcare professionals. Various predictive models have been developed. Compared with traditional machine learning models, deep learning-based models have achieved more promising performance. However, due to the lack of transparency, the acceptance of deep learning-based models are often limited. This paper proposes a Stacked Attention-based Network, SANet, for accurate and interpretable health risk prediction. Two novel attention-based modules, named Convolutional Attention Module and Sequential Attention Module respectively, are designed to capture patient-specific contextual information at both feature and sequence levels. Particularly, Sequential Attention Module can flexibly learn the impact of the time interval between sequential visits and significantly enhance the interpretability and robustness of learning outcomes from sequences. Experimental results on two real-world EHR datasets demonstrate the superior predictive accuracy of our method, as well as interpretability and robustness, compared to existing state-of-the-art methods. The findings extracted by this approach are also empirically confirmed by relevant literature and medical experts.

Predicting the health risks of patients using Electronic Health Records (EHR) has attracted considerable attention in recent years, especially with the development of deep learning techniques. Health risk refers to the probability of the occurrence of a specific health outcome for a specific patient. The predicted risks can be used to support decision-making by healthcare professionals. EHRs are structured patient journey data. Each patient journey contains a chronological set of clinical events, and within each clinical event, there is a set of clinical/medical activities. Due to variations of patient conditions and treatment needs, EHR patient journey data has an inherently high degree of missingness that contains important information affecting relationships among variables, including time. Existing deep learning-based models generate imputed values for missing values when learning the relationships. However, imputed data in EHR patient journey data may distort the clinical meaning of the original EHR patient journey data, resulting in classification bias. This paper proposes a novel end-to-end approach to modeling EHR patient journey data with Integrated Convolutional and Recurrent Neural Networks. Our model can capture both long- and short-term temporal patterns within each patient journey and effectively handle the high degree of missingness in EHR data without any imputation data generation. Extensive experimental results using the proposed model on two real-world datasets demonstrate robust performance as well as superior prediction accuracy compared to existing state-of-the-art imputation-based prediction methods.

Electronic Health Records (EHRs) exhibit a high amount of missing data due to variations of patient conditions and treatment needs. Imputation of missing values has been considered an effective approach to deal with this challenge. Existing work separates imputation method and prediction model as two independent parts of an EHR-based machine learning system. We propose an integrated end-to-end approach by utilizing a Compound Density Network (CDNet) that allows the imputation method and prediction model to be tuned together within a single framework. CDNet consists of a Gated recurrent unit (GRU), a Mixture Density Network (MDN), and a Regularized Attention Network (RAN). The GRU is used as a latent variable model to model EHR data. The MDN is designed to sample latent variables generated by GRU. The RAN serves as a regularizer for less reliable imputed values. The architecture of CDNet enables GRU and MDN to iteratively leverage the output of each other to impute missing values, leading to a more accurate and robust prediction. We validate CDNet on the mortality prediction task on the MIMIC-III dataset. Our model outperforms state-of-the-art models by significant margins. We also empirically show that regularizing imputed values is a key factor for superior prediction performance. Analysis of prediction uncertainty shows that our model can capture both aleatoric and epistemic uncertainties, which offers model users a better understanding of the model results.