TrajVis: a visual clinical decision support system to translate artificial intelligence trajectory models in the precision management of chronic kidney disease.

OBJECTIVE: Our objective is to develop and validate TrajVis, an interactive tool that assists clinicians in using artificial intelligence (AI) models to leverage patients' longitudinal electronic medical records (EMRs) for personalized precision management of chronic disease progression. MATERIALS AND METHODS: We first perform requirement analysis with clinicians and data scientists to determine the visual analytics tasks of the TrajVis system as well as its design and functionalities. A graph AI model for chronic kidney disease (CKD) trajectory inference named DisEase PrOgression Trajectory (DEPOT) is used for system development and demonstration. TrajVis is implemented as a full-stack web application with synthetic EMR data derived from the Atrium Health Wake Forest Baptist Translational Data Warehouse and the Indiana Network for Patient Care research database. A case study with a nephrologist and a user experience survey of clinicians and data scientists are conducted to evaluate the TrajVis system. RESULTS: The TrajVis clinical information system is composed of 4 panels: the Patient View for demographic and clinical information, the Trajectory View to visualize the DEPOT-derived CKD trajectories in latent space, the Clinical Indicator View to elucidate longitudinal patterns of clinical features and interpret DEPOT predictions, and the Analysis View to demonstrate personal CKD progression trajectories. System evaluations suggest that TrajVis supports clinicians in summarizing clinical data, identifying individualized risk predictors, and visualizing patient disease progression trajectories, overcoming the barriers of AI implementation in healthcare. DISCUSSION: The TrajVis system provides a novel visualization solution which is complimentary to other risk estimators such as the Kidney Failure Risk Equations. CONCLUSION: TrajVis bridges the gap between the fast-growing AI/ML modeling and the clinical use of such models for personalized and precision management of chronic diseases.

TrialView: An AI-powered Visual Analytics System for Temporal Event Data in Clinical Trials.

Randomized controlled trials (RCT) are the gold standards for evaluating the efficacy and safety of therapeutic interventions in human subjects. In addition to the pre-specified endpoints, trial participants' experience reveals the time course of the intervention. Few analytical tools exist to summarize and visualize the individual experience of trial participants. Visual analytics allows integrative examination of temporal event patterns of patient experience, thus generating insights for better care decisions. Towards this end, we introduce TrialView, an information system that combines graph artificial intelligence (AI) and visual analytics to enhance the dissemination of trial data. TrialView offers four distinct yet interconnected views: Individual, Cohort, Progression, and Statistics, enabling an interactive exploration of individual and group-level data. The TrialView system is a general-purpose analytical tool for a broad class of clinical trials. The system is powered by graph AI, knowledge-guided clustering, explanatory modeling, and graph-based agglomeration algorithms. We demonstrate the system's effectiveness in analyzing temporal event data through a case study.

ARDaC Common Data Model Facilitates Data Dissemination and Enables Data Commons for Modern Clinical Studies.

Modern clinical studies collect longitudinal and multimodal data about participants, treatments and responses, biospecimens, and molecular and multiomics data. Such rich and complex data requires new common data models (CDM) to support data dissemination and research collaboration. We have developed the ARDaC CDM for the Alcoholic Hepatitis Network (AlcHepNet) Research Data Commons (ARDaC) to support clinical studies and translational research in the national AlcHepNet consortium. The ARDaC CDM bridges the gap between the data models used by the AlcHepNet electronic data capture platform (REDCap) and the Genomic Data Commons (GDC) data model used by the Gen3 data commons framework. It extends the GDC data model for clinical studies; facilitates the harmonization of research data across consortia and programs; and supports the development of the ARDaC. ARDaC CDM is designed as a general and extensible CDM for addressing the needs of modern clinical studies. The ARDaC CDM is available at https://dev.ardac.org/DD.

SpaRx: elucidate single-cell spatial heterogeneity of drug responses for personalized treatment.

Spatial cellular authors heterogeneity contributes to differential drug responses in a tumor lesion and potential therapeutic resistance. Recent emerging spatial technologies such as CosMx, MERSCOPE and Xenium delineate the spatial gene expression patterns at the single cell resolution. This provides unprecedented opportunities to identify spatially localized cellular resistance and to optimize the treatment for individual patients. In this work, we present a graph-based domain adaptation model, SpaRx, to reveal the heterogeneity of spatial cellular response to drugs. SpaRx transfers the knowledge from pharmacogenomics profiles to single-cell spatial transcriptomics data, through hybrid learning with dynamic adversarial adaption. Comprehensive benchmarking demonstrates the superior and robust performance of SpaRx at different dropout rates, noise levels and transcriptomics coverage. Further application of SpaRx to the state-of-the-art single-cell spatial transcriptomics data reveals that tumor cells in different locations of a tumor lesion present heterogenous sensitivity or resistance to drugs. Moreover, resistant tumor cells interact with themselves or the surrounding constituents to form an ecosystem for drug resistance. Collectively, SpaRx characterizes the spatial therapeutic variability, unveils the molecular mechanisms underpinning drug resistance and identifies personalized drug targets and effective drug combinations.

SiGra: single-cell spatial elucidation through an image-augmented graph transformer.

Recent advances in high-throughput molecular imaging have pushed spatial transcriptomics technologies to subcellular resolution, which surpasses the limitations of both single-cell RNA-seq and array-based spatial profiling. The multichannel immunohistochemistry images in such data provide rich information on the cell types, functions, and morphologies of cellular compartments. In this work, we developed a method, single-cell spatial elucidation through image-augmented Graph transformer (SiGra), to leverage such imaging information for revealing spatial domains and enhancing substantially sparse and noisy transcriptomics data. SiGra applies hybrid graph transformers over a single-cell spatial graph. SiGra outperforms state-of-the-art methods on both single-cell and spot-level spatial transcriptomics data from complex tissues. The inclusion of immunohistochemistry images improves the model performance by 37% (95% CI: 27-50%). SiGra improves the characterization of intratumor heterogeneity and intercellular communication and recovers the known microscopic anatomy. Overall, SiGra effectively integrates different spatial modality data to gain deep insights into spatial cellular ecosystems.

DEPOT: graph learning delineates the roles of cancers in the progression trajectories of chronic kidney disease using electronic medical records.

Chronic kidney disease (CKD) is a common, complex, and heterogeneous disease impacting aging populations. Determining the landscape of disease progression trajectories from midlife to senior age in a real-world context allows us to better understand the progression of CKD, the heterogeneity of progression patterns among the risk population, and the interactions with other clinical conditions like cancers. In this study, we use electronic health records (EHRs) to outline the CKD progression trajectory roadmap for the Wake Forest Baptist Medical Center (WFBMC) patient population. We establish an EHR cohort (n = 79,434) with patients' health status identified by 18 Essential Clinical Indices across 508,732 clinical encounters. We develop the DisEase PrOgression Trajectory (DEPOT) approach to model CKD progression trajectories and individualize clinical decision support. The DEPOT is an evidence-driven, graph-based clinical informatics approach that addresses the unique challenges in longitudinal EHR data by systematically using the graph artificial intelligence (graph-AI) model for representation learning and reverse graph embedding for trajectory reconstruction. Moreover, DEPOT includes a prediction model to assign new patients along the progression trajectory. We successfully establish the EHR-based CKD progression trajectories with DEPOT in the WFUBMC cohort. We annotate the trajectories with clinical features, including kidney function, age, and other indices, including cancer. This CKD progression trajectory roadmap reveals diverse kidney failure pathways associated with different clinical conditions. Specifically, we have identified one high-risk trajectory and two low-risk trajectories. Switching pathways from low-risk trajectories to the high-risk one is associated with accelerated decline in kidney function. On this roadmap, high-risk patients are enriched in the skin and GU cancers, which differs from low-risk patients, suggesting fundamentally different disease progression mechanisms. Overall, the CKD progression trajectory roadmap reveals novel diverse renal failure pathways in type 2 diabetes mellitus and highlights disease progression patterns associated with cancer phenotypes.

SpaRx: Elucidate single-cell spatial heterogeneity of drug responses for personalized treatment.

Spatial cellular heterogeneity contributes to differential drug responses in a tumor lesion and potential therapeutic resistance. Recent emerging spatial technologies such as CosMx SMI, MERSCOPE, and Xenium delineate the spatial gene expression patterns at the single cell resolution. This provides unprecedented opportunities to identify spatially localized cellular resistance and to optimize the treatment for individual patients. In this work, we present a graph-based domain adaptation model, SpaRx, to reveal the heterogeneity of spatial cellular response to drugs. SpaRx transfers the knowledge from pharmacogenomics profiles to single-cell spatial transcriptomics data, through hybrid learning with dynamic adversarial adaption. Comprehensive benchmarking demonstrates the superior and robust performance of SpaRx at different dropout rates, noise levels, and transcriptomics coverage. Further application of SpaRx to the state-of-art single-cell spatial transcriptomics data reveals that tumor cells in different locations of a tumor lesion present heterogenous sensitivity or resistance to drugs. Moreover, resistant tumor cells interact with themselves or the surrounding constituents to form an ecosystem for drug resistance. Collectively, SpaRx characterizes the spatial therapeutic variability, unveils the molecular mechanisms underpinning drug resistance, and identifies personalized drug targets and effective drug combinations. Key Points: We have developed a novel graph-based domain adaption model named SpaRx, to reveal the heterogeneity of spatial cellular response to different types of drugs, which bridges the gap between pharmacogenomics knowledgebase and single-cell spatial transcriptomics data.SpaRx is developed tailored for single-cell spatial transcriptomics data and is provided available as a ready-to-use open-source software, which demonstrates high accuracy and robust performance.SpaRx uncovers that tumor cells located in different areas within tumor lesion exhibit varying levels of sensitivity or resistance to drugs. Moreover, SpaRx reveals that tumor cells interact with themselves and the surrounding microenvironment to form an ecosystem capable of drug resistance.