Search

Satellite imagery depicts the Earth's surface remotely and provides comprehensive information for many applications, such as land use monitoring and urban planning. Existing studies on unsupervised representation learning for satellite images only take into account the images' geographic information, ignoring human activity factors. To bridge this gap, we propose using the Point-of-Interest (POI) data to capture human factors and designing a contrastive learning-based framework to consolidate the representation of satellite imagery with POI information. Besides, we introduce a season-invariant representation learning model on satellite imagery, considering that human factors are mostly unchanging with respect to seasons. An attention model is designed at last to merge the representations from the geographic, seasonal, and POI perspectives adaptively. On the basis of real-world datasets collected from Beijing,1 we evaluate our method for predicting socioeconomic indicators. The results show that the representation containing POI information outperforms the geographic representation in estimating commercial activity-related indicators. Our proposed attentional framework can estimate the socioeconomic indicators with R2 of 0.874 and outperforms the baseline methods. Furthermore, we explore the differences in the representations of satellite images with varying socioeconomic statuses. Finally, we investigate the impact of geographic and POI perspective information in the representation learning process, as well as the effect of satellite imagery on various spatial resolutions. [ABSTRACT FROM AUTHOR]

Familial hemophagocytic lymphohistiocytosis Type 2 (FHL2) associated central nervous system (CNS) involvement is less understood in children, especially when considering neurologic manifestations as part of the initial presentation. We conducted a retrospective review of the clinical manifestations and genetic abnormality of four Han Chinese children with FHL2 who were patients at the neurology department of Beijing Children's Hospital from November 2015 to October 2018. These four patients initially manifested CNS symptoms in their disease presentation, and all four patients were misdiagnosed as having ademyelinating disease, such as acute disseminated encephalomyelitis and multiple sclerosis. Given these misdiagnoses, it is important that general physicians and pediatricians maintain awareness of the possibility of FHL2 as a differential diagnosis. These four cases included neurologic manifestations including seizures, ataxia, spasticity, gait disorder, and coma. Bilateral abnormal signals in the cerebrum, including in white matter, gray matter, and junctions were discovered. Enhanced magnetic resonance imaging (MRI) in these patients showed spot or ring enhancement and/or hemorrhage. These patients all possessed a compound heterozygote mutation PRF1 gene. Whole exome sequencing analysis revealed seven different mutations (three novel mutations) spread over the PRF1 gene and a heterozygous missense mutation c.1349C > T [p.T450M] that was present in two patients. Three novel mutations, c.634T > C[p.Y212H], c.1083_1094del[p.361_364del], and c.1306G > T [p.D436Y], were discovered and through in silico analysis were discovered to be deleterious. Neurologic manifestations were the initial symptoms of FHL2 in these patients in addition to the expected leukopenia and hepatosplenomegaly. Whole exome sequencing of PRF1 for patients with similar presentations would facilitate prompt and accurate diagnosis and treatment. [ABSTRACT FROM AUTHOR]

In the past decade, due to the management policies and coal combustion controls in Beijing, the consumption of natural gas has increased gradually. Nevertheless, the research on the emission characteristics of gaseous pollutants emitted from gas-fired industrial boilers, especially considering the influence of low nitrogen (low-NOx) retrofit policy of gas boilers, is scarcely. In this study, based on literature and field investigations, onsite measurements of NOx, CO, NH 3 and VOCs (Volatile Organic Compounds) emissions from gas-fired industrial boilers as well as the key factors that affected the emission of gaseous pollutants were discussed. Category-specific emission factors (EFs) of NOx, CO, NH 3 and VOCs were obtained from the field measurements of 1107 “low-NOx” retrofitted and unabated gas-fired industrial boilers. Our results showed that operating load and control measures were the two key factors affecting the formation of gaseous pollutants. The EFs of NOx (EF NOx ) and CO (EF CO ) of atmospheric combustion boilers (ACBs) were much higher than the EFs of chamber combustion boilers (CCBs). The total emissions of NOx, CO, NH 3 and VOCs from gas-fired industrial boilers in Beijing in the year of 2015 were estimated at 10489.6 t, 3272.8 t, 196.4 t and 235.4 t, respectively. Alkanes, BTEX, oxygenated VOCs and non-reactive organic matter were the four main chemical components of VOCs. As for the spatial distributions, the emissions of NOx, CO, NH 3 and VOCs from gas-fired industrial boilers in Beijing were predominantly concentrated in central six urban districts. In the future, more detailed investigation and field tests for all kinds of gas-fired industrial boilers are still greatly needed to achieve more reliable estimations of atmospheric pollutants from gas-fired industrial boilers. [ABSTRACT FROM AUTHOR]