Cloud-Based Machine Learning Platform to Predict Clinical Outcomes at Home for Patients With Cardiovascular Conditions Discharged From Hospital: Clinical Trial.

BACKGROUND: Hospitalizations account for almost one-third of the US $4.1 trillion health care cost in the United States. A substantial portion of these hospitalizations are attributed to readmissions, which led to the establishment of the Hospital Readmissions Reduction Program (HRRP) in 2012. The HRRP reduces payments to hospitals with excess readmissions. In 2018, >US $700 million was withheld; this is expected to exceed US $1 billion by 2022. More importantly, there is nothing more physically and emotionally taxing for readmitted patients and demoralizing for hospital physicians, nurses, and administrators. Given this high uncertainty of proper home recovery, intelligent monitoring is needed to predict the outcome of discharged patients to reduce readmissions. Physical activity (PA) is one of the major determinants for overall clinical outcomes in diabetes, hypertension, hyperlipidemia, heart failure, cancer, and mental health issues. These are the exact comorbidities that increase readmission rates, underlining the importance of PA in assessing the recovery of patients by quantitative measurement beyond the questionnaire and survey methods. OBJECTIVE: This study aims to develop a remote, low-cost, and cloud-based machine learning (ML) platform to enable the precision health monitoring of PA, which may fundamentally alter the delivery of home health care. To validate this technology, we conducted a clinical trial to test the ability of our platform to predict clinical outcomes in discharged patients. METHODS: Our platform consists of a wearable device, which includes an accelerometer and a Bluetooth sensor, and an iPhone connected to our cloud-based ML interface to analyze PA remotely and predict clinical outcomes. This system was deployed at a skilled nursing facility where we collected >17,000 person-day data points over 2 years, generating a solid training database. We used these data to train our extreme gradient boosting (XGBoost)-based ML environment to conduct a clinical trial, Activity Assessment of Patients Discharged from Hospital-I, to test the hypothesis that a comprehensive profile of PA would predict clinical outcome. We developed an advanced data-driven analytic platform that predicts the clinical outcome based on accurate measurements of PA. Artificial intelligence or an ML algorithm was used to analyze the data to predict short-term health outcome. RESULTS: We enrolled 52 patients discharged from Stanford Hospital. Our data demonstrated a robust predictive system to forecast health outcome in the enrolled patients based on their PA data. We achieved precise prediction of the patients' clinical outcomes with a sensitivity of 87%, a specificity of 79%, and an accuracy of 85%. CONCLUSIONS: To date, there are no reliable clinical data, using a wearable device, regarding monitoring discharged patients to predict their recovery. We conducted a clinical trial to assess outcome data rigorously to be used reliably for remote home care by patients, health care professionals, and caretakers.

PAX4 loss of function increases diabetes risk by altering human pancreatic endocrine cell development.

The coding variant (p.Arg192His) in the transcription factor PAX4 is associated with an altered risk for type 2 diabetes (T2D) in East Asian populations. In mice, Pax4 is essential for beta cell formation but its role on human beta cell development and/or function is unknown. Participants carrying the PAX4 p.His192 allele exhibited decreased pancreatic beta cell function compared to homozygotes for the p.192Arg allele in a cross-sectional study in which we carried out an intravenous glucose tolerance test and an oral glucose tolerance test. In a pedigree of a patient with young onset diabetes, several members carry a newly identified p.Tyr186X allele. In the human beta cell model, EndoC-ßH1, PAX4 knockdown led to impaired insulin secretion, reduced total insulin content, and altered hormone gene expression. Deletion of PAX4 in human induced pluripotent stem cell (hiPSC)-derived islet-like cells resulted in derepression of alpha cell gene expression. In vitro differentiation of hiPSCs carrying PAX4 p.His192 and p.X186 risk alleles exhibited increased polyhormonal endocrine cell formation and reduced insulin content that can be reversed with gene correction. Together, we demonstrate the role of PAX4 in human endocrine cell development, beta cell function, and its contribution to T2D-risk.

Zmiz1 is required for mature ß-cell function and mass expansion upon high fat feeding.

OBJECTIVE: Identifying the transcripts which mediate genetic association signals for type 2 diabetes (T2D) is critical to understand disease mechanisms. Studies in pancreatic islets support the transcription factor ZMIZ1 as a transcript underlying a T2D GWAS signal, but how it influences T2D risk is unknown. METHODS: ß-Cell-specific Zmiz1 knockout (Zmiz1ßKO) mice were generated and phenotypically characterised. Glucose homeostasis was assessed in Zmiz1ßKO mice and their control littermates on chow diet (CD) and high fat diet (HFD). Islet morphology and function were examined by immunohistochemistry and in vitro islet function was assessed by dynamic insulin secretion assay. Transcript and protein expression were assessed by RNA sequencing and Western blotting. In islets isolated from genotyped human donors, we assessed glucose-dependent insulin secretion and islet insulin content by static incubation assay. RESULTS: Male and female Zmiz1ßKO mice were glucose intolerant with impaired insulin secretion, compared with control littermates. Transcriptomic profiling of Zmiz1ßKO islets identified over 500 differentially expressed genes including those involved in ß-cell function and maturity, which we confirmed at the protein level. Upon HFD, Zmiz1ßKO mice fail to expand ß-cell mass and become severely diabetic. Human islets from carriers of the ZMIZ1-linked T2D-risk alleles have reduced islet insulin content and glucose-stimulated insulin secretion. CONCLUSIONS: ß-Cell Zmiz1 is required for normal glucose homeostasis. Genetic variation at the ZMIZ1 locus may influence T2D-risk by reducing islet mass expansion upon metabolic stress and the ability to maintain a mature ß-cell state.

Large and Small Animal Models of Heart Failure With Reduced Ejection Fraction.

Heart failure (HF) describes a heterogenous complex spectrum of pathological conditions that results in structural and functional remodeling leading to subsequent impairment of cardiac function, including either systolic dysfunction, diastolic dysfunction, or both. Several factors chronically lead to HF, including cardiac volume and pressure overload that may result from hypertension, valvular lesions, acute, or chronic ischemic injuries. Major forms of HF include hypertrophic, dilated, and restrictive cardiomyopathy. The severity of cardiomyopathy can be impacted by other comorbidities such as diabetes or obesity and external stress factors. Age is another major contributor, and the number of patients with HF is rising worldwide in part due to an increase in the aged population. HF can occur with reduced ejection fraction (HF with reduced ejection fraction), that is, the overall cardiac function is compromised, and typically the left ventricular ejection fraction is lower than 40%. In some cases of HF, the ejection fraction is preserved (HF with preserved ejection fraction). Animal models play a critical role in facilitating the understanding of molecular mechanisms of how hearts fail. This review aims to summarize and describe the strengths, limitations, and outcomes of both small and large animal models of HF with reduced ejection fraction that are currently used in basic and translational research. The driving defect is a failure of the heart to adequately supply the tissues with blood due to impaired filling or pumping. An accurate model of HF with reduced ejection fraction would encompass the symptoms (fatigue, dyspnea, exercise intolerance, and edema) along with the pathology (collagen fibrosis, ventricular hypertrophy) and ultimately exhibit a decrease in cardiac output. Although countless experimental studies have been published, no model completely recapitulates the full human disease. Therefore, it is critical to evaluate the strength and weakness of each animal model to allow better selection of what animal models to use to address the scientific question proposed.

The Right Heart Network and Risk Stratification in Pulmonary Arterial Hypertension.

BACKGROUND: Prognosis in pulmonary arterial hypertension (PAH) is closely related to indexes of right ventricular function. A better understanding of their relationship may provide important implications for risk stratification in PAH. RESEARCH QUESTION: Can clinical network graphs inform risk stratification in PAH? STUDY DESIGN AND METHODS: The study cohort consisted of 231 patients with PAH followed up for a median of 7.1 years. An undirected, correlation network was used to visualize the relationship between clinical features in PAH. This network was enriched for right heart parameters and included N-terminal pro-hormone B-type natriuretic peptide (NT-proBNP), comprehensive echocardiographic parameters, and hemodynamics, as well as 6-min walk distance (6MWD), vital signs, laboratory data, and diffusing capacity for carbon monoxide (Dlco). Connectivity was assessed by using eigenvector and betweenness centrality to reflect global and regional connectivity, respectively. Cox proportional hazards regression was used to model event-free survival for the combined end point of death or lung transplantation. RESULTS: A network of closely intertwined features centered around NT-proBNP with 6MWD emerging as a secondary hub were identified. Less connected nodes included Dlco, systolic BP, albumin, and sodium. Over the follow-up period, death or transplantation occurred in 92 patients (39.8%). A strong prognostic model was achieved with a Harrell's C-index of 0.81 (0.77-0.85) when combining central right heart features (NT-proBNP and right ventricular end-systolic remodeling index) with 6MWD and less connected nodes (Dlco, systolic BP, albumin, sodium, sex, connective tissue disease etiology, and prostanoid therapy). When added to the baseline risk model, serial change in NT-proBNP significantly improved outcome prediction at 5 years (increase in C-statistic of 0.071 ± 0.024; P = .003). INTERPRETATION: NT-proBNP emerged as a central hub in the intertwined PAH network. Connectivity analysis provides explainability for feature selection and combination in outcome models.

Evolution of Cytomegalovirus-Responsive T Cell Clonality following Solid Organ Transplantation.

CMV infection is a significant complication after solid organ transplantation. We used single cell TCR αß sequencing to determine how memory inflation impacts clonality and diversity of the CMV-responsive CD8 and CD4 T cell repertoire in the first year after transplantation in human subjects. We observed CD8 T cell inflation but no changes in clonal diversity, indicating homeostatic stability in clones. In contrast, the CD4 repertoire was diverse and stable over time, with no evidence of CMV-responsive CD4 T cell expansion. We identified shared CDR3 TCR motifs among patients but no public CMV-specific TCRs. Temporal changes in clonality in response to transplantation and in the absence of detectable viral reactivation suggest changes in the repertoire immediately after transplantation followed by an expansion with stable clonal competition that may mediate protection.

Integrative metabolomics and transcriptomics identifies itaconate as an adjunct therapy to treat ocular bacterial infection.

The eye is highly susceptible to inflammation-mediated tissue damage evoked during bacterial infection. However, mechanisms regulating inflammation to protect the eye remain elusive. Here, we used integrated metabolomics and transcriptomics to show that the immunomodulatory metabolite itaconate and immune-responsive gene 1 (Irg1) are induced in bacterial (Staphylococcus aureus)-infected mouse eyes, bone-marrow-derived macrophages (BMDMs), and Müller glia. Itaconate levels are also elevated in the vitreous of patients with bacterial endophthalmitis. Irg1 deficiency in mice led to increased ocular pathology. Conversely, intraocular administration of itaconate protects both Irg1-/- and wild-type mice from bacterial endophthalmitis by reducing inflammation, bacterial burden, and preserving retinal architecture and visual function. Notably, itaconate exerts synergistic effects with antibiotics. The protective, anti-inflammatory effects of itaconate are mediated via activation of NRF2/HO-1 signaling and inhibition of NLRP3 inflammasome. Collectively, our study demonstrates the Irg1/itaconate axis is a regulator of intraocular inflammation and provides evidence for using itaconate, along with antibiotics, to treat bacterial infections.

Alteration in ventricular pressure stimulates cardiac repair and remodeling.

The mammalian heart undergoes complex structural and functional remodeling to compensate for stresses such as pressure overload. While studies suggest that, at best, the adult mammalian heart is capable of very limited regeneration arising from the proliferation of existing cardiomyocytes, how myocardial stress affects endogenous cardiac regeneration or repair is unknown. To define the relationship between left ventricular afterload and cardiac repair, we induced left ventricle pressure overload in adult mice by constriction of the ascending aorta (AAC). One week following AAC, we normalized ventricular afterload in a subset of animals through removal of the aortic constriction (de-AAC). Subsequent monitoring of cardiomyocyte cell cycle activity via thymidine analog labeling revealed that an acute increase in ventricular afterload induced cardiomyocyte proliferation. Intriguingly, a release in ventricular overload (de-AAC) further increases cardiomyocyte proliferation. Following both AAC and de-AAC, thymidine analog-positive cardiomyocytes exhibited characteristics of newly generated cardiomyocytes, including single diploid nuclei and reduced cell size as compared to age-matched, sham-operated adult mouse myocytes. Notably, those smaller cardiomyocytes frequently resided alongside one another, consistent with local stimulation of cellular proliferation. Collectively, our data demonstrate that adult cardiomyocyte proliferation can be locally stimulated by an acute increase or decrease of ventricular pressure, and this mode of stimulation can be harnessed to promote cardiac repair.

Determination of system level alterations in host transcriptome due to Zika virus (ZIKV) Infection in retinal pigment epithelium.

Previously, we reported that Zika virus (ZIKV) causes ocular complications such as chorioretinal atrophy, by infecting cells lining the blood-retinal barrier, including the retinal pigment epithelium (RPE). To understand the molecular basis of ZIKV-induced retinal pathology, we performed a meta-analysis of transcriptome profiles of ZIKV-infected human primary RPE and other cell types infected with either ZIKV or other related flaviviruses (Japanese encephalitis, West Nile, and Dengue). This led to identification of a unique ZIKV infection signature comprising 43 genes (35 upregulated and 8 downregulated). The major biological processes perturbed include SH3/SH2 adaptor activity, lipid and ceramide metabolism, and embryonic organ development. Further, a comparative analysis of some differentially regulated genes (ABCG1, SH2B3, SIX4, and TNFSF13B) revealed that ZIKV induced their expression relatively more than dengue virus did in RPE. Importantly, the pharmacological inhibition of ABCG1, a membrane transporter of cholesterol, resulted in reduced ZIKV infectivity. Interestingly, the ZIKV infection signature revealed the downregulation of ALDH5A1 and CHML, genes implicated in neurological (cognitive impairment, expressive language deficit, and mild ataxia) and ophthalmic (choroideremia) disorders, respectively. Collectively, our study revealed that ZIKV induces differential gene expression in RPE cells, and the identified genes/pathways (e.g., ABCG1) could potentially contribute to ZIKV-associated ocular pathologies.

Towards precision medicine: discovering novel gynecological cancer biomarkers and pathways using linked data.

BACKGROUND: Next Generation Sequencing (NGS) is playing a key role in therapeutic decision making for the cancer prognosis and treatment. The NGS technologies are producing a massive amount of sequencing datasets. Often, these datasets are published from the isolated and different sequencing facilities. Consequently, the process of sharing and aggregating multisite sequencing datasets are thwarted by issues such as the need to discover relevant data from different sources, built scalable repositories, the automation of data linkage, the volume of the data, efficient querying mechanism, and information rich intuitive visualisation. RESULTS: We present an approach to link and query different sequencing datasets (TCGA, COSMIC, REACTOME, KEGG and GO) to indicate risks for four cancer types - Ovarian Serous Cystadenocarcinoma (OV), Uterine Corpus Endometrial Carcinoma (UCEC), Uterine Carcinosarcoma (UCS), Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma (CESC) - covering the 16 healthy tissue-specific genes from Illumina Human Body Map 2.0. The differentially expressed genes from Illumina Human Body Map 2.0 are analysed together with the gene expressions reported in COSMIC and TCGA repositories leading to the discover of potential biomarkers for a tissue-specific cancer. CONCLUSION: We analyse the tissue expression of genes, copy number variation (CNV), somatic mutation, and promoter methylation to identify associated pathways and find novel biomarkers. We discovered twenty (20) mutated genes and three (3) potential pathways causing promoter changes in different gynaecological cancer types. We propose a data-interlinked platform called BIOOPENER that glues together heterogeneous cancer and biomedical repositories. The key approach is to find correspondences (or data links) among genetic, cellular and molecular features across isolated cancer datasets giving insight into cancer progression from normal to diseased tissues. The proposed BIOOPENER platform enriches mutations by filling in missing links from TCGA, COSMIC, REACTOME, KEGG and GO datasets and provides an interlinking mechanism to understand cancer progression from normal to diseased tissues with pathway components, which in turn helped to map mutations, associated phenotypes, pathways, and mechanism.