Diabetic retinopathy (DR) is a major reason for the increased visual loss globally, and it became an important cause of visual impairment among people in 25-74 years of age. The DR significantly ...affects the economic status in society, particularly in healthcare systems. When timely treatment is provided to the DR patients, approximately 90% of patients can be saved from visual loss. Therefore, it becomes highly essential to classify the stages and severity of DR for the recommendation of required treatments. In this view, this paper introduces a new automated Hyperparameter Tuning Inception-v4 (HPTI-v4) model for the detection and classification of DR from color fundus images. At the preprocessing stage, the contrast level of the fundus image will be improved by the use of contrast limited adaptive histogram equalization (CLAHE) model. Then, the segmentation of the preprocessed image takes place utilizing a histogram-based segmentation model. Afterward, the HPTI-v4 model is applied to extract the required features from the segmented image and it subsequently undergoes classification by the use of a multilayer perceptron (MLP). A series of experiments take place on MESSIDOR (Methods to Evaluate Segmentation and Indexing Techniques in the field of Retinal Ophthalmology) DR dataset to guarantee the goodness of the HPTI-v4 approach and the obtained results clearly exhibited the supremacy of the HPTI-v4 model over the compared methods in a significant way.
Abstract
Seasonally, the East Asian particulate matter (PM) level is higher in the winter–spring period than in summer, at which time the level rapidly decreases due to the summer monsoon migration. ...Attempting to attribute East Asian PM pollution to a source without considering such natural factors is challenging. However, to what degree the effect of season on an attribution bias remains controversial; the bias may even be implicated in PM-related health effects. This study examined seasonal dynamics including the unusual precipitation evolution during 2020—a year in which coronavirus-related lockdowns occurred frequently worldwide—and suggested a large-scale effect from the removal of PM pollutants from most of the coastal cities in East Asia. In winter–spring 2020, compared with that of previous years, a deeper and farther southward intrusion of the East Asian coastal trough and a stronger surface monsoon flow acted jointly to transport air pollutants over the Korea–Japan region. In summer 2020, the strength and migration of the western North Pacific (WNP) high increased precipitation and removed air pollutants in mid-latitude East Asia, whereas it reduced precipitation in the subtropical WNP. Consequently, the reduced PM level in the subtropical region (including Taiwan) may be irrelevant to the anomalous seasonal pattern. Although an artificial effect is conceivable and may be primarily responsible for the marked decrease in 2020 East Asian PM pollutants in some subtropical cities, the modulation of a large-scale and precipitating effect also deserves consideration.
In order to generate a probability density function (PDF) for fitting the probability distributions of practical data, this study proposes a deep learning method which consists of two stages: (1) a ...training stage for estimating the cumulative distribution function (CDF) and (2) a performing stage for predicting the corresponding PDF. The CDFs of common probability distributions can be utilised as activation functions in the hidden layers of the proposed deep learning model for learning actual cumulative probabilities, and the differential equation of the trained deep learning model can be used to estimate the PDF. Numerical experiments with single and mixed distributions are conducted to evaluate the performance of the proposed method. The experimental results show that the values of both CDF and PDF can be precisely estimated by the proposed method.
This study proposes a mobile positioning method that adopts recurrent neural network algorithms to analyze the received signal strength indications from heterogeneous networks (e.g., cellular ...networks and Wi-Fi networks) for estimating the locations of mobile stations. The recurrent neural networks with multiple consecutive timestamps can be applied to extract the features of time series data for the improvement of location estimation. In practical experimental environments, there are 4525 records, 59 different base stations, and 582 different Wi-Fi access points detected in Fuzhou University in China. The lower location errors can be obtained by the recurrent neural networks with multiple consecutive timestamps (e.g., two timestamps and three timestamps); from the experimental results, it can be observed that the average error of location estimation was 9.19 m by the proposed mobile positioning method with two timestamps.
Change in extreme events in climate projections is a major concern. If the frequency of dry events is expected to increase in a warmer climate (thus, the overall number of wet days will decrease), ...heavy and extreme precipitation are also expected to increase because of a shift of the precipitation spectrum. However, the forecasts exhibit numerous uncertainties.
This study focuses on the Asian region, separated into the following three subregions: the East Asian region, the Indian region, and western North Pacific region, where the summer monsoon can bring heavy rainfall. Particularly emphasized herein is the reliability of the projection, using data from a large ensemble of 30 models from phase 5 of the Coupled Model Intercomparison Project. The scattering of the ensemble enables obtaining an optimal estimate of the uncertainties, and it is used to compute the correlation between projected changes of extreme events and circulation changes.
The results show clear spatial and temporal variations in the confidence of changes, with results being more reliable during the wet season (i.e., the summer monsoon). The ensemble predicts changes in atmospheric circulation with favorable confidence, especially in the low-level moisture flux convergence (MFC). However, the correlation between this mean change and the modification of extreme events is nonsignificant. Also analyzed herein are the correlation and change of MFC exclusively during these events. The horizontal MFC exerts a nonnegligible influence on the change in the intensity of extremes. However, it is mostly the change in vertical circulation and moisture advection that is correlated with the change in frequency and intensity of extreme events.
Celotno besedilo
Dostopno za:
BFBNIB, DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
We report on molecular dynamics simulation evidence revealing that an oligomer additive can be used to greatly facilitate the self-assembly of a bisurea in organic solvent media, through the initial ...regular packing and the subsequent stiffening of the self-assembly filament. The underlying physics is attributed to the substantially reduced diffusivities of the solute and, in particular, solvent molecules, featuring a generally weakened (thermal) Brownian force under ambient conditions. Without such oligomer-induced molecular cooling-in contrast to the usual external cooling, the original solvent medium is noted to foster instead more stabilized and disordered aggregates and, in particular, it would require a temperature reduction that is practically inaccessible in order to sustain similar stiffness of the self-assembly filament. These features, in accord with recent experimental observations, highlight the open opportunity of promoting the self-assembly of small functional molecules in general solvent media without requiring substantial changes of the system temperature, as is crucial for many practical applications including the biological/biomedical ones.
Oligomer-induced molecular cooling of the solvent medium helps to foster a highly ordered and stiff bisurea self-assembly filament.
Graph embedding is an effective yet efficient way to convert graph data into a low dimensional space. In recent years, deep learning has applied on graph embedding and shown outstanding performance. ...Adjacency matrix is often taken as the storage data structure of graph. However, there are the problems of insufficient spatial proximity information in adjacency matrix. Therefore, this study proposes a deep community detection method which includes (1) matrix reconstruction method, (2) spatial feature extraction method and (3) community detection method. The original adjacency matrix in social network is reconstructed based on the opinion leader and nearer neighbors for obtaining spatial proximity matrix. The spatial proximity matrix can obtain subspace of the graph which can help convolution neural network easily and quickly extract the spatial localization. The spatial eigenvector of reconstructed adjacency matrix can be extracted by an auto-encoder based on convolution neural network for the improvement of modularity. In experiments, four open datasets of practical social networks were selected to evaluate the proposed method, and the experimental results show that the proposed deep community detection method obtained higher modularity than other deep learning methods. Therefore, the proposed deep community detection method can effectively detect high quality communities in social networks.
The Asian Monsoonal rainfall accounts for the majority of annual regional precipitation in East and South Asia and could be remotely regulated by El Niño‐Southern Oscillation (ENSO). Besides, several ...paleoclimate records and simulations have indicated solar signals in the Asian Monsoon, implying the impact of solar activity on the regional monsoon precipitation. By conducting multi‐linear regression analysis to the solar irradiance forced single‐forcing experiment in the last millennium, this study presents the comparison of solar and ENSO effects on monsoonal precipitation in South and East Asia during early summer (May–June). Increased total solar irradiance during high solar activity years tends to trigger a favorable environment for developing monsoon onset, leading to more precipitation against ENSO‐related patterns over Southeast and South Asia before peak‐summer (July–August). The result supports reconstructed terrestrial records and underlines considerable influences of the solar cycle on the variation of the Asian Summer Monsoon.
Plain Language Summary
Previous studies have shown the influence of the solar cycle on various climate systems, such as El Niño‐Southern Oscillation (ENSO), Atlantic Oscillation, Pacific Decadal Oscillation, etc., despite the relatively small amount of solar irradiance variations to the total amount of solar insolation. In addition, paleo‐records indicate the correlation between the long‐term solar cycle and the variation of the Asian Monsoon. This study attempts to clarify the role of solar activity in monsoonal precipitation over South and East Asia. We find that during early summer (May–June), high solar activity tends to enhance the meridional temperature gradient over the northern Indian Ocean. An associated meridional circulation is thereby intensified and well‐coupled with monsoonal circulations. The coupling contributes to more moisture transport and increased precipitation over the subcontinent, contrasting with the reduced rainfall and suppressed monsoonal circulation over the northern Indian Ocean during warmer ENSO events. This result indicates that the effects of changes in solar irradiation on climate systems should be considered, particularly in the Asian Monsoon region. Furthermore, a physical explanation is provided for the solar signals in the Asian Monsoon in paleo‐evidences.
Key Points
The Multi‐Linear Regression analysis is used to isolate solar activity and El Niño‐Southern Oscillation (ENSO) signals from the CESM‐LME project's solar‐only experiment
Sub‐seasonal analyses to the Asian Summer Monsoon are highlighted by varied responses to solar activity in early and peak summer
Higher solar activity increases early summer rainfall around Southeast Asia against ENSO‐related patterns by enhancing monsoonal circulation
The summer rainfall climate of East Asia underwent large and abrupt changes during past climates, in response to precessional forcing, glacial–interglacial cycles as well as abrupt changes to the ...North Atlantic during the Last Glacial. However, current interpretations of said changes are typically formulated in terms of modulation of summer monsoon intensity, and do not account for the known complexity in the seasonal evolution of East Asian rainfall, which exhibits sharp transition from the Spring regime to the Meiyu, and then again from the Meiyu to the Summer regime.
We explore the interpretation that East Asian rainfall climate undergoes a modulation of its seasonality during said paleoclimate changes. Following previous suggestions we focus on role of the westerly jet over Asia, namely that its latitude relative to Tibet is critical in determining the stepwise transitions in East Asian rainfall seasons. In support of this linkage, we show from observational data that the interannual co-variation of June (July–August) rainfall and upper tropospheric zonal winds show properties consistent with an altered timing of the transition to the Meiyu (Summer), and with more northward-shifted westerlies for earlier transitions.
We similarly suggest that East Asian paleoclimate changes resulted from an altered timing in the northward evolution of the jet and hence the seasonal transitions, in particular the transition of the jet from south of the Plateau to the north that determines the seasonal transition from Spring rains to the Meiyu. In an extreme scenario – which we speculate the climate system tended towards during stadial (cold) phases of D/O stadials and periods of low Northern Hemisphere summer insolation – the jet does not jump north of the Plateau, essentially keeping East Asia in prolonged Spring conditions.
We argue that this hypothesis provides a viable explanation for a key paleoproxy signature of D/O stadials over East Asia, namely the heavier mean δ18O of precipitation as recorded in speleothem records. The southward jet position prevents the low-level monsoonal flow – which is isotopically light – from penetrating into the interior of East Asia; as such, precipitation there will be heavier, consistent with speleothem records. This hypothesis can also explain other key evidences of East Asian paleoclimate changes, in particular the occurrence of dusty conditions during North Atlantic stadials, and the southward migration of the Holocene optimal rainfall.
•We explore role of seasonal rainfall transitions in East Asian paleoclimate change.•Seasonal regimes determined by meridional position of westerlies relative to Tibet.•East Asian paleoclimate changes reflect systematic meridional shifts to westerlies.•Modern-day analogs and model simulations support this hypothesis.•Hypothesis may partly explain cave records of East Asian paleoclimate.
Westerly perturbation is enlarged over the Far East‐Okhotsk region in late June and early July and is associated with the largest land‐sea heating contrast surrounding the Sea of Okhotsk. The ...corresponding characteristics in the lower troposphere are southward deepening of the cold low over northeastern China, and intensification of the Okhotsk high. Coincidentally, the Meiyu‐Baiu coupled with the western North Pacific (WNP) subtropical high is nearly stagnant during this period. By simulations using a global climate model with an intensified Okhotsk surface high in response to cooling Sea of Okhotsk, it is suggested that the enhanced thermal contrast over the Far East‐Okhotsk region can generate an obstacle to Meiyu‐Baiu poleward migration. Corresponding to the intensified Okhotsk high, the WNP subtropical high is strengthened by the high ridge over Taiwan. The East Asian midlatitude westerly jet stream in the northern flank of the WNP subtropical high also strengthens; the consequently enhanced midtropospheric warm temperature advection can regulate the latitude position of the Meiyu‐Baiu. The wave source generated in the upper troposphere is located over the midlatitude WNP (anomalous cyclone) bordering the Sea of Okhotsk, whereas that in the lower troposphere is related to the strengthened westerly in the northern WNP subtropical high. The wave activity propagation consistently indicates the strengthening and equatorward confinement of the westerly jet. Therefore, the intensified Okhotsk high and enlarged westerly perturbation beginning in late June are suggested as an inherently geographic limit of the Far East‐Okhotsk region in regard to Meiyu‐Baiu migration.
Key Points
The Okhotsk high and surrounding westerly perturbation strengthen in late June, coinciding with the near stagnant nature of the Meiyu‐Baiu
Experiment of cooling Sea of Okhotsk confirms the effect of intensified Okhotsk high on fixing the western North Pacific subtropical high
An inherent geographic limit of the Far East‐Okhotsk region in regard to Meiyu‐Baiu migration is suggested