Sleep Classification With Artificial Synthetic Imaging Data Using Convolutional Neural Networks.

OBJECTIVE: We propose a new analytic framework, "Artificial Synthetic Imaging Data (ASID) Workflow," for sleep classification from a wearable device comprising: 1) the creation of ASID from data collected by a non-invasive wearable device that permits real-time multi-modal physiological monitoring on heart rate (HR), 3-axis accelerometer, electrodermal activity, and skin temperature, denoted as "Temporal E4 Data" (TED) and 2) the use of an image classification supervised learning algorithm, convolutional neural network (CNN), to classify periods of sleep. METHODS: We investigate ASID Workflow under 6 settings (3 data resolutions × 2 HR scenarios). Competing machine/deep learning classification algorithms, including logistic regression, support vector machine, random forest, k-nearest neighbors, and Long Short-Term Memory, are applied to TED as comparisons, termed "Competing Workflow." RESULTS: The ASID Workflow achieves excellent performance with mean weighted accuracy across settings of 94.7%, and is superior to the Competing Workflow with high and low resolution data regardless of the inclusion of HR modality. This superiority is maximized for low resolution data without HR. Additionally, CNN has a relatively low subject-wise test computational cost compared with competing algorithms. CONCLUSION: We demonstrate the utility of creating ASID from multi-modal physiological data and applying a preexisting image classification algorithm to achieve better classification accuracy. We shed light on the influence of data resolution and HR modality on the Workflow's performance. SIGNIFICANCE: Applying CNN to ASID allows us to capture both temporal and spatial dependency among physiological variables and modalities by using 2D images' topological structure that competing algorithms fail to utilize.

Predicting temporal propagation of seasonal influenza using improved gaussian process model.

Influenza rapidly spreads in seasonal epidemics and imposes a considerable economic burden on hospitals and other healthcare costs. Thus, predicting the propagation of influenza accurately is crucial in preventing influenza outbreaks and protecting public health. Most current studies focus on the spread simulation of influenza. However, few studies have investigated the dependencies between meteorological variables and influenza activity. This study develops a non-parametric model based on Gaussian process regression for influenza prediction considering meteorological effect to capture temporal dependencies hidden in influenza time series. To identify the most explanatory external variables, L1-regularization is applied to identify meteorology factor subsets, and three types of covariance functions are designed to characterize non-stationary and periodic behavior in influenza activity. The dependencies of diseases and meteorology are modeled through the designed cross-covariance function. A real case in Shenzhen, China was studied to validate our proposed model along with comparisons to recently developed multivariate statistical models for influenza prediction. Results show that our proposed influenza prediction approach achieves superior performance in terms of one-week-ahead prediction of influenza-like illness.

The Association between Air Pollution and Outpatient and Inpatient Visits in Shenzhen, China.

Nowadays, air pollution is a severe environmental problem in China. To investigate the effects of ambient air pollution on health, a time series analysis of daily outpatient and inpatient visits in 2015 were conducted in Shenzhen (China). Generalized additive model was employed to analyze associations between six air pollutants (namely SO2, CO, NO2, O3, PM10, and PM2.5) and daily outpatient and inpatient visits after adjusting confounding meteorological factors, time and day of the week effects. Significant associations between air pollutants and two types of hospital visits were observed. The estimated increase in overall outpatient visits associated with each 10 µg/m³ increase in air pollutant concentration ranged from 0.48% (O3 at lag 2) to 11.48% (SO2 with 2-day moving average); for overall inpatient visits ranged from 0.73% (O3 at lag 7) to 17.13% (SO2 with 8-day moving average). Our results also suggested a heterogeneity of the health effects across different outcomes and in different populations. The findings in present study indicate that even in Shenzhen, a less polluted area in China, significant associations exist between air pollution and daily number of overall outpatient and inpatient visits.

Coordinative modulation of human zinc transporter 2 gene expression through active and suppressive regulators.

Zinc transporter 2 (ZnT2) is one of the cellular factors responsible for Zn homeostasis. Upon Zn overload, ZnT2 reduces cellular Zn by transporting it into excretory vesicles. We investigated the molecular mechanism that regulates human ZnT2 (hZnT2) gene expression. Zn induces hZnT2 expression in dose- and time-dependent manners. Overexpression of metal-responsive transcription factor 1 (MTF-1) increases hZnT2 transcription, whereas depletion of MTF-1 reduces hZnT2 expression. There are five putative metal response elements (MREs) within 1kb upstream of the hZnT2 gene. A serial deletion of the hZnT2 promoter region (from 5' to 3') shows that the two MREs proximal to the gene are essential for Zn-induced promoter activity. Further mutation analysis concludes that the penultimate MRE (MREb) supports the metal-induced promoter activity. The hZnT2 promoter has also a zinc finger E-box binding homeobox (ZEB) binding element. Mutation or deletion of this ZEB binding element elevates the basal and Zn-induced hZnT2 promoter activities. Knockdown of ZEB1 mRNA enhances the hZnT2 transcript level in HEK-293 cells. In MCF-7 (ZEB-deficient) cells, expression of ZEB proteins attenuates the Zn-induced hZnT2 expression. However, expressions of MTF-1 target genes such as human ZnT1 and metallothionein IIA were not affected. Our study shows the expression of the hZnT2 gene is coordinately regulated via active and suppressive modulators.

Expression and characterization of SUMO-conjugated metal-responsive transcription factor 1: SIM-dependent cross-interaction and distinct DNA binding activity.

Metal-responsive transcription factor 1 (MTF-1) regulates a variety of genes involving in metal homeostasis and oxidative stress. We have shown that MTF-1 can be conjugated by small ubiquitin-like modifier (SUMO) and forms complexes with cellular factor(s) in a SUMO-interacting motif (SIM)-dependent manner. To investigate whether the interaction of MTF-1 and its SUMO conjugate occurs, we expressed and isolated MTF-1 and sumoylated MTF-1 (S-MTF-1) for functional studies. Various conditions were examined to optimize the expressions of MTF-1 and S-MTF-1. Results from affinity column chromatography demonstrated that the unmodified MTF-1 consistently co-eluted with the S-MTF-1. Mutations at the SIM did not reduce the level of MTF-1 sumoylation but the sumoylated product can then be purified to homogeneity. The presence of MTF-1 cross-interaction was further supported by in vitro pull-down assays. The ability of the purified proteins in binding metal-responsive element (MRE) was assessed with electrophoretic mobility shift assay. Noticeably, MTF-1 required the presence of cell extracts to render the binding activity. However, S-MTF-1 binds MRE in void of other cellular factors. The same characteristic was found for MTF-1 with SUMO fusion at the carboxyl terminus. These results indicate that the presence of SUMO moiety allows the protein to interact directly with MRE.

PTEN interacts with metal-responsive transcription factor 1 and stimulates its transcriptional activity.

MTF-1 (metal-responsive transcription factor 1) is an essential mammalian protein for embryonic development and modulates the expression of genes involving in zinc homoeostasis and responding to oxidative stress. We report in the present paper that PTEN (phosphatase and tensin homologue deleted on chromosome 10) associates with MTF-1 in the cells. These two proteins interact via the acidic domain of MTF-1 and the phosphatase/C2 domain of PTEN. Depletion of PTEN reduced MT (metallothionein) gene expression and increased cellular sensitivity to cadmium toxicity. PTEN did not alter the nuclear translocation, protein stability or DNA-binding activity of MTF-1. Zinc increased MTF-1-PTEN interaction in a dose-dependent manner. The interaction elevated within 2 h of zinc addition and declined afterwards in the cells. The enhanced binding activity occurred mainly in the cytoplasm and reduced after translocating the MTF-1 into the nucleus. Blocking signalling through the PI3K (phosphoinositide 3-kinase) pathway did not alter the zinc-induced MT expression. Analysis of enzymatically inactive PTEN mutants demonstrated that protein but not lipid phosphatase activity of PTEN was involved in the regulation of MTF-1 activity. The same regulatory role of PTEN was also noted in the regulation of ZnT1 (zinc transporter 1), another target gene of MTF-1.

The role of small ubiquitin-like modifier-interacting motif in the assembly and regulation of metal-responsive transcription factor 1.

Metal-responsive transcription factor 1 (MTF-1) is an essential protein required for mouse embryonic development. We report here the occurrence of sumoylation on MTF-1. Mutational studies demonstrated that sumoylation occurs on the lysine residue at position 627 (Lys(627)) of mouse MTF-1. Small ubiquitin-like modifier (SUMO)-1 was fused to the C terminus of MTF-1 to mimic the sumoylated form of the protein and it suppressed the transcriptional activity of MTF-1. The nuclear translocation activity, DNA-binding activity, and protein stability of SUMO-fused MTF-1 are similar to that of wild type MTF-1. The level of sumoylation was reduced by metal in a dose- and time-dependent manner. The fact that zinc reduces MTF-1 sumoylation makes the suppressive role of sumoylated MTF-1 in transcription physiologically less significant because the SUMO moiety of MTF-1 is removed when MTF-1 translocates into nucleus. We further identified a SUMO-interacting motif (SIM) on MTF-1. Remarkably, MTF-1 binds sumoylated MTF-1 and/or other cellular factors in a SIM-dependent manner. This interaction was disrupted by treating cells with zinc. Gel permeation chromatography demonstrated that MTF-1 forms SIM-dependent complexes. This cross-interaction transpires in the cytoplasm and markedly reduces upon nuclear translocation. It can therefore be concluded that SUMO conjugation and the SIM on MTF-1 do not play a critical role in suppressing transcriptional activity. Instead, MTF-1 forms complexes with cellular factors through SIM and SUMO moiety in the cytoplasm. The result explores a new understanding for the mode of MTF-1 assembly and regulation in cells.