What dictates income in New York City? SHAP analysis of income estimation based on Socio-economic and Spatial Information Gaussian Processes (SSIG).

Income inequality presents a key challenge to urban sustainability across the developed economies. Traditionally, accurate high granularity income data are generally obtained from field surveys. However, due to privacy considerations, field subjects are hesitant to provide accurate personal income data. A Socio-economic & Spatial-Information-GP (SSIG) model is thereby developed to estimate district-based high granularity income for New York City (NYC). As compared to the state-of-the-art Gaussian Processes (GP) income estimation model based entirely on spatial information, SSIG incorporates socio-economic domain-specific knowledge into a GP model. For SSIG to be explainable, SHapley Additive exPlanations (SHAP) analysis is undertaken to evaluate the relative contribution of various key individual socio-economic variables to district-based per-capita and median household income in NYC. Differentiating from traditional income inequality studies based predominantly on linear or log-linear regression model, SSIG presents a novel income-based model architecture, capable of modelling complex non-linear relationships. In parallel, SHAP analysis serves an effective analytical tool for identifying the key attributes to income inequality. Results have shown that SSIG surpasses other state-of-the-art baselines in estimation accuracy, as far as per-capita and median household income estimation at the Tract-level and the ZIP-level in NYC are concerned. SHAP results have indicated that having a bachelor or a postgraduate degree can accurately predict income in NYC, despite that between-district income inequality due to Sex/Race remains prevalent. SHAP has further confirmed that between-district income gap is more associated with Race than Sex. Furthermore, ablation study shows that socio-economic information is more predictive of income at the ZIP-level, relative to the spatial information. This study carries significant implications for policy-making in a developed context. To promote urban economic sustainability in NYC, policymakers can attend to the growing income disparity (income inequality) contributed by Sex and Race, while giving more higher education opportunities to residents in the lower-income districts, as the estimated per-capita income is more sensitive to the proportion of adults ≥25 holding a bachelor's degree. Finally, interpretative SHAP analysis is useful for investigating the relative contribution of socio-economic inputs to any predicted outputs in future machine-learning-driven socio-economic analyses.

Somatic Mutations and Alzheimer's Disease.

Alzheimer's disease (AD) represents a global health challenge, with an estimated 55 million people suffering from the non-curable disease across the world. While amyloid-ß plaques and tau neurofibrillary tangles in the brain define AD proteinopathy, it has become evident that diverse coding and non-coding regions of the genome may significantly contribute to AD neurodegeneration. The diversity of factors associated with AD pathogenesis, coupled with age-associated damage, suggests that a series of triggering events may be required to initiate AD. Since somatic mutations accumulate with aging, and aging is a major risk factor for AD, there is a great potential for somatic mutational events to drive disease. Indeed, recent data from the Gozes team/laboratories as well as other leading laboratories correlated the accumulation of somatic brain mutations with the progression of tauopathy. In this review, we lay the current perspectives on the principal genetic factors associated with AD and the potential causes, highlighting the contribution of somatic mutations to the pathogenesis of late onset Alzheimer's disease. The roles that artificial intelligence and big data can play in accelerating the progress of causal somatic mutation markers/biomarkers identification, and the associated drug discovery/repurposing, have been highlighted for future AD and other neurodegenerations, with the aim to bring hope for the vulnerable aging population.

Oral microbiota signatures in post-traumatic stress disorder (PTSD) veterans.

Post-traumatic stress disorder (PTSD) represents a global public health concern, affecting about 1 in 20 individuals. The symptoms of PTSD include intrusiveness (involuntary nightmares or flashbacks), avoidance of traumatic memories, negative alterations in cognition and mood (such as negative beliefs about oneself or social detachment), increased arousal and reactivity with irritable reckless behavior, concentration problems, and sleep disturbances. PTSD is also highly comorbid with anxiety, depression, and substance abuse. To advance the field from subjective, self-reported psychological measurements to objective molecular biomarkers while considering environmental influences, we examined a unique cohort of Israeli veterans who participated in the 1982 Lebanon war. Non-invasive oral 16S RNA sequencing was correlated with psychological phenotyping. Thus, a microbiota signature (i.e., decreased levels of the bacteria sp_HMT_914, 332 and 871 and Noxia) was correlated with PTSD severity, as exemplified by intrusiveness, arousal, and reactivity, as well as additional psychopathological symptoms, including anxiety, hostility, memory difficulties, and idiopathic pain. In contrast, education duration correlated with significantly increased levels of sp_HMT_871 and decreased levels of Bacteroidetes and Firmicutes, and presented an inverted correlation with adverse psychopathological measures. Air pollution was positively correlated with PTSD symptoms, psychopathological symptoms, and microbiota composition. Arousal and reactivity symptoms were correlated with reductions in transaldolase, an enzyme controlling a major cellular energy pathway, that potentially accelerates aging. In conclusion, the newly discovered bacterial signature, whether an outcome or a consequence of PTSD, could allow for objective soldier deployment and stratification according to decreases in sp_HMT_914, 332, 871, and Noxia levels, coupled with increases in Bacteroidetes levels. These findings also raise the possibility of microbiota pathway-related non-intrusive treatments for PTSD.

A novel deep learning-based 3D cell segmentation framework for future image-based disease detection.

Cell segmentation plays a crucial role in understanding, diagnosing, and treating diseases. Despite the recent success of deep learning-based cell segmentation methods, it remains challenging to accurately segment densely packed cells in 3D cell membrane images. Existing approaches also require fine-tuning multiple manually selected hyperparameters on the new datasets. We develop a deep learning-based 3D cell segmentation pipeline, 3DCellSeg, to address these challenges. Compared to the existing methods, our approach carries the following novelties: (1) a robust two-stage pipeline, requiring only one hyperparameter; (2) a light-weight deep convolutional neural network (3DCellSegNet) to efficiently output voxel-wise masks; (3) a custom loss function (3DCellSeg Loss) to tackle the clumped cell problem; and (4) an efficient touching area-based clustering algorithm (TASCAN) to separate 3D cells from the foreground masks. Cell segmentation experiments conducted on four different cell datasets show that 3DCellSeg outperforms the baseline models on the ATAS (plant), HMS (animal), and LRP (plant) datasets with an overall accuracy of 95.6%, 76.4%, and 74.7%, respectively, while achieving an accuracy comparable to the baselines on the Ovules (plant) dataset with an overall accuracy of 82.2%. Ablation studies show that the individual improvements in accuracy is attributable to 3DCellSegNet, 3DCellSeg Loss, and TASCAN, with the 3DCellSeg demonstrating robustness across different datasets and cell shapes. Our results suggest that 3DCellSeg can serve a powerful biomedical and clinical tool, such as histo-pathological image analysis, for cancer diagnosis and grading.

Outdoor PM2.5 concentration and rate of change in COVID-19 infection in provincial capital cities in China.

This study investigates thoroughly whether acute exposure to outdoor PM2.5 concentration, P, modifies the rate of change in the daily number of COVID-19 infections (R) across 18 high infection provincial capitals in China, including Wuhan. A best-fit multiple linear regression model was constructed to model the relationship between P and R, from 1 January to 20 March 2020, after accounting for meteorology, net move-in mobility (NM), time trend (T), co-morbidity (CM), and the time-lag effects. Regression analysis shows that P (ß = 0.4309, p < 0.001) is the most significant determinant of R. In addition, T (ß = -0.3870, p < 0.001), absolute humidity (AH) (ß = 0.2476, p = 0.002), P × AH (ß = -0.2237, p < 0.001), and NM (ß = 0.1383, p = 0.003) are more significant determinants of R, as compared to GDP per capita (ß = 0.1115, p = 0.015) and CM (Asthma) (ß = 0.1273, p = 0.005). A matching technique was adopted to demonstrate a possible causal relationship between P and R across 18 provincial capital cities. A 10 µg/m3 increase in P gives a 1.5% increase in R (p < 0.001). Interaction analysis also reveals that P × AH and R are negatively correlated (ß = -0.2237, p < 0.001). Given that P exacerbates R, we recommend the installation of air purifiers and improved air ventilation to reduce the effect of P on R. Given the increasing observation that COVID-19 is airborne, measures that reduce P, plus mandatory masking that reduces the risks of COVID-19 associated with viral-particulate transmission, are strongly recommended. Our study is distinguished by the focus on the rate of change instead of the individual cases of COVID-19 when modelling the statistical relationship between R and P in China; causal instead of correlation analysis via the matching analysis, while taking into account the key confounders, and the individual plus the interaction effects of P and AH on R.

Demographic effects on facial emotion expression: an interdisciplinary investigation of the facial action units of happiness.

Understanding demographic difference in facial expression of happiness has crucial implications on social communication. However, prior research on facial emotion expression has mostly focused on the effect of a single demographic factor (typically gender, race, or age), and is limited by the small image dataset collected in laboratory settings. First, we used 30,000 (4800 after pre-processing) real-world facial images from Flickr, to analyze the facial expression of happiness as indicated by the intensity level of two distinctive facial action units, the Cheek Raiser (AU6) and the Lip Corner Puller (AU12), obtained automatically via a deep learning algorithm that we developed, after training on 75,000 images. Second, we conducted a statistical analysis on the intensity level of happiness, with both the main effect and the interaction effect of three core demographic factors on AU12 and AU6. Our results show that females generally display a higher AU12 intensity than males. African Americans tend to exhibit a higher AU6 and AU12 intensity, when compared with Caucasians and Asians. The older age groups, especially the 40-69-year-old, generally display a stronger AU12 intensity than the 0-3-year-old group. Our interdisciplinary study provides a better generalization and a deeper understanding on how different gender, race and age groups express the emotion of happiness differently.

Putative Blood Somatic Mutations in Post-Traumatic Stress Disorder-Symptomatic Soldiers: High Impact of Cytoskeletal and Inflammatory Proteins.

BACKGROUND: We recently discovered autism/intellectual disability somatic mutations in postmortem brains, presenting higher frequency in Alzheimer's disease subjects, compared with the controls. We further revealed high impact cytoskeletal gene mutations, coupled with potential cytoskeleton-targeted repair mechanisms. OBJECTIVE: The current study was aimed at further discerning if somatic mutations in brain diseases are presented only in the most affected tissue (the brain), or if blood samples phenocopy the brain, toward potential diagnostics. METHODS: Variant calling analyses on an RNA-seq database including peripheral blood samples from 85 soldiers (58 controls and 27 with symptoms of post-traumatic stress disorder, PTSD) was performed. RESULTS: High (e.g., protein truncating) as well as moderate impact (e.g., single amino acid change) germline and putative somatic mutations in thousands of genes were found. Further crossing the mutated genes with autism, intellectual disability, cytoskeleton, inflammation, and DNA repair databases, identified the highest number of cytoskeletal-mutated genes (187 high and 442 moderate impact). Most of the mutated genes were shared and only when crossed with the inflammation database, more putative high impact mutated genes specific to the PTSD-symptom cohorts versus the controls (14 versus 13) were revealed, highlighting tumor necrosis factor specifically in the PTSD-symptom cohorts. CONCLUSION: With microtubules and neuro-immune interactions playing essential roles in brain neuroprotection and Alzheimer-related neurodegeneration, the current mutation discoveries contribute to mechanistic understanding of PTSD and brain protection, as well as provide future diagnostics toward personalized military deployment strategies and drug design.

Disturbance-Aware Neuro-Optimal System Control Using Generative Adversarial Control Networks.

Disturbance, which is generally unknown to the controller, is unavoidable in real-world systems and it may affect the expected system state and output. Existing control methods, like robust model predictive control, can produce robust solutions to maintain the system stability. However, these robust methods trade the solution optimality for stability. In this article, a method called generative adversarial control networks (GACNs) is proposed to train a controller via demonstrations of the optimal controller. By formulating the optimal control problem in the presence of disturbance, the controller trained by GACNs obtains neuro-optimal solutions without knowing the future disturbance and determines the objective function explicitly. A joint loss, composed of the adversarial loss and the least square loss, is designed to be used in the training of the generator. Experimental results on simulated systems with disturbance show that GACNs outperform other compared control methods.

A Revisit of Infinite Population Models for Evolutionary Algorithms on Continuous Optimization Problems.

Infinite population models are important tools for studying population dynamics of evolutionary algorithms. They describe how the distributions of populations change between consecutive generations. In general, infinite population models are derived from Markov chains by exploiting symmetries between individuals in the population and analyzing the limit as the population size goes to infinity. In this article, we study the theoretical foundations of infinite population models of evolutionary algorithms on continuous optimization problems. First, we show that the convergence proofs in a widely cited study were in fact problematic and incomplete. We further show that the modeling assumption of exchangeability of individuals cannot yield the transition equation. Then, in order to analyze infinite population models, we build an analytical framework based on convergence in distribution of random elements which take values in the metric space of infinite sequences. The framework is concise and mathematically rigorous. It also provides an infrastructure for studying the convergence of the stacking of operators and of iterating the algorithm which previous studies failed to address. Finally, we use the framework to prove the convergence of infinite population models for the mutation operator and the k-ary recombination operator. We show that these operators can provide accurate predictions for real population dynamics as the population size goes to infinity, provided that the initial population is identically and independently distributed.

Multi-Type Sensor Placements in Gaussian Spatial Fields for Environmental Monitoring.

As citizens are increasingly concerned about the surrounding environment, it is important for modern cities to provide sufficient and accurate environmental information to the public for decision making in the era of smart cities. Due to the limited budget, we often need to optimize the sensor placement in order to maximize the overall information gain according to certain criteria. Existing work is primarily concerned with single-type sensor placement; however, the environment usually requires accurate measurements of multiple types of environmental characteristics. In this paper, we focus on the optimal multi-type sensor placement in Gaussian spatial field for environmental monitoring. We study two representative cases: the one-with-all case when each station is equipped with all types of sensors and the general case when each station is equipped with at least one type of sensor. We propose two greedy algorithms accordingly, each with a provable approximation guarantee. We evaluated the proposed approach via an application in air quality monitoring scenario in Hong Kong and experimental results demonstrate the effectiveness of the proposed approach.

A review on health cost accounting of air pollution in China.

Over the last three decades, rapid industrialization in China has generated an unprecedentedly high level of air pollution and associated health problems. Given that China accounts for one-fifth of the world population and suffers from severe air pollution, a comprehensive review of the indicators accounting for the health costs in relation to air pollution will benefit evidence-based and health-related environmental policy-making. This paper reviews the conventional static and the new dynamic approach adopted for air pollution-related health cost accounting in China and analyzes the difference between the two in estimating GDP loss. The advantages of adopting the dynamic approach for health cost accounting in China, with conditions guaranteeing its optimal performance are highlighted. Guidelines on how one can identify an appropriate approach for health cost accounting in China are put forward. Further, we outline and compare the globally-applicable and China-specific indicators adopted by different accounting methodologies, with their pros and cons being discussed. A comprehensive account of the available databases and methodologies for health cost accounting in China are outlined. Future directions to guide health cost accounting in China are provided. Our work provides valuable insights into future health cost accounting research in China. Our study has strengthen the view that the dynamic approach is comparatively more preferred than the static approach for health cost accounting in China, if more data is available to train the dynamic models and improve the robustness of the parameters employed. In addition, future dynamic model should address the socio-economic impacts, including benefits or losses of air pollution polices, to provide a more robust policy picture. Our work has laid the key principles and guidelines for selecting proper econometric approaches and parameters. We have also identified a proper estimation method for the Value of Life in China, and proposed the integration of engineering approaches, such as the use of deep learning and big data analysis for health cost accounting at the fine-grained level (city-district or sub-regional level). Our work has also identified the gap for more accurate health cost accounting at the fine-grained level in China, which will subsequently affect the quality of health-related air pollution policy decision-making at such levels, and the health-related quality of life of the citizens in China.

Rate-splitting multiple access for downlink communication systems: bridging, generalizing, and outperforming SDMA and NOMA.

Space-division multiple access (SDMA) utilizes linear precoding to separate users in the spatial domain and relies on fully treating any residual multi-user interference as noise. Non-orthogonal multiple access (NOMA) uses linearly precoded superposition coding with successive interference cancellation (SIC) to superpose users in the power domain and relies on user grouping and ordering to enforce some users to fully decode and cancel interference created by other users. In this paper, we argue that to efficiently cope with the high throughput, heterogeneity of quality of service (QoS), and massive connectivity requirements of future multi-antenna wireless networks, multiple access design needs to depart from those two extreme interference management strategies, namely fully treat interference as noise (as in SDMA) and fully decode interference (as in NOMA). Considering a multiple-input single-output broadcast channel, we develop a novel multiple access framework, called rate-splitting multiple access (RSMA). RSMA is a more general and more powerful multiple access for downlink multi-antenna systems that contains SDMA and NOMA as special cases. RSMA relies on linearly precoded rate-splitting with SIC to decode part of the interference and treat the remaining part of the interference as noise. This capability of RSMA to partially decode interference and partially treat interference as noise enables to softly bridge the two extremes of fully decoding interference and treating interference as noise and provides room for rate and QoS enhancements and complexity reduction. The three multiple access schemes are compared, and extensive numerical results show that RSMA provides a smooth transition between SDMA and NOMA and outperforms them both in a wide range of network loads (underloaded and overloaded regimes) and user deployments (with a diversity of channel directions, channel strengths, and qualities of channel state information at the transmitter). Moreover, RSMA provides rate and QoS enhancements over NOMA at a lower computational complexity for the transmit scheduler and the receivers (number of SIC layers).

A stratified acoustic model accounting for phase shifts for underwater acoustic networks.

Accurate acoustic channel models are critical for the study of underwater acoustic networks. Existing models include physics-based models and empirical approximation models. The former enjoy good accuracy, but incur heavy computational load, rendering them impractical in large networks. On the other hand, the latter are computationally inexpensive but inaccurate since they do not account for the complex effects of boundary reflection losses, the multi-path phenomenon and ray bending in the stratified ocean medium. In this paper, we propose a Stratified Acoustic Model (SAM) based on frequency-independent geometrical ray tracing, accounting for each ray's phase shift during the propagation. It is a feasible channel model for large scale underwater acoustic network simulation, allowing us to predict the transmission loss with much lower computational complexity than the traditional physics-based models. The accuracy of the model is validated via comparisons with the experimental measurements in two different oceans. Satisfactory agreements with the measurements and with other computationally intensive classical physics-based models are demonstrated.

Improved short adjacent repeat identification using three evolutionary Monte Carlo schemes.

This paper employs three Evolutionary Monte Carlo (EMC) schemes to solve the Short Adjacent Repeat Identification Problem (SARIP), which aims to identify the common repeat units shared by multiple sequences. The three EMC schemes, i.e., Random Exchange (RE), Best Exchange (BE), and crossover are implemented on a parallel platform. The simulation results show that compared with the conventional Markov Chain Monte Carlo (MCMC) algorithm, all three EMC schemes can not only shorten the computation time via speeding up the convergence but also improve the solution quality in difficult cases. Moreover, we observe that the performances of different EMC schemes depend on the degeneracy degree of the motif pattern.