High Quality Region Research Articles

The mechanism and path of pollution reduction and carbon reduction affecting high quality economic development - taking the Yangtze River Delta urban agglomeration as an example

In the wake of the COVID19 pandemic, what is crucial for the continued promotion of ecological civilization construction and sustainable green development in China is to achieve the dynamic and balanced development between pollution reduction, carbon emission reduction, and economic growth. Therefore, the indexes of CO2 emissions, pollution reduction intensity, and high-quality economic development are quantified in this study by using the panel data related to 26 cities in the Yangtze River Delta urban agglomeration(YRDUA) from 2010 to 2020. Furthermore, kernel density analysis, GIS spatial analysis, and other methods are employed to demonstrate the spatial and temporal patterns of pollution reduction intensity, CO2 emission levels, and the characteristics of high-quality economic development in the YRDUA. In order to reveal the dynamic relationship between pollution reduction, carbon emission reduction, and high-quality economic development, a panel vector autoregressive model (PAVR) was established and the system generalized method of moments was used to estimate the model coefficients. Additionally, Granger causality tests, impulse response analysis, and variance decomposition were conducted. There is a fluctuating upward trend shown by the overall level of CO2 emissions in the YRDUA, with significant spatial nonuniformity exhibited in carbon emissions. Despite the self-promotion mechanism playing a more significant role in the high-quality economic development in the YRDUA and its three regions, the impact of pollutant and carbon emissions on high-quality economic development is quite limited. A significant promoting effect is exhibited in those high and moderately high-quality economic development regions, while nonlinear effects are observed in those high and moderately high-quality economic development regions. As for the interactive relationships, the three factors are correlated with each other to a more significant extent in those high and moderately high-quality economic development regions, which contributes to mutual influence and causal relationship. However, there is no effective interactive mechanism observed among the three factors in the whole area and low-quality economic development regions, with high-quality economic development playing a major role in environmental pollution and carbon emission reduction.

SA3Det: Detecting Rotated Objects via Pixel-Level Attention and Adaptive Labels Assignment

Remote sensing of rotated objects often encounters numerous small and dense objects. To tackle small-object neglect and inaccurate angle predictions in elongated objects, we propose SA3Det, a novel method employing Pixel-Level Attention and Adaptive Labels Assignment. First, we introduce a self-attention module that learns dense pixel-level relations between features extracted by the backbone and neck, effectively preserving and exploring the spatial relationships of potential small objects. We then introduce an adaptive label assignment strategy that refines proposals by assigning labels based on loss, enhancing sample selection during training. Additionally, we designed an angle-sensitive module that enhances angle prediction by learning rotational feature maps and incorporating multi-angle features. These modules significantly enhance detection accuracy and yield high-quality region proposals. Our approach was validated by experiments on the DOTA and HRSC2016 datasets, demonstrating that SA3Det achieves mAPs of 76.31% and 89.4%, respectively.

Computer Vision for E-learning Student Face Reaction Detection for Improving Learning Rate

In the advent of e-learning, understanding student engagement and reaction is crucial for improving the quality of education and enhancing the learning rate. With the advancement of computer vision technologies, there is a significant opportunity to analyze and interpret student reactions in a non-intrusive manner. This study proposes a novel framework employing Faster R-CNN integrated with DenseNet architecture for real-time detection of student facial reactions during e-learning sessions. The proposed method leverages the strengths of Faster R-CNN in generating high-quality region proposals for object detection tasks, coupled with the DenseNet’s efficiency in feature propagation and reduction in the number of parameters, which is well-suited for processing the intricate patterns in facial expressions. Our approach begins with the application of Faster R-CNN to extract potential facial regions with high accuracy and reduced computational cost. The integration of DenseNet as a backbone for feature extraction within Faster R-CNN capitalizes on its densely connected convolutional networks, ensuring maximum information flow between layers in the network. By doing so, the system becomes exceptionally adept at recognizing subtle changes in facial features that indicate various student reactions, such as confusion, engagement, or boredom. We conducted a series of experiments using a diverse dataset of e-learning interactions, collected under various lighting conditions and involving multiple ethnicities to ensure robustness and generalizability. The model was trained and validated on this dataset, and the results demonstrate a significant improvement in detection rates of student reactions compared to existing methods.

Interferometric phase diagram generation method based on dot trace texture information and semantic segmentation of correlation coefficients

This paper proposes a method for generating interferometric phase diagrams, which includes interferometric image registration based on Dot Trace Texture Information (DTTI) and interferometric phase filtering based on semantic segmentation. First, the InSAR images are sharpened according to the different backscattering characteristics of various pixel values, and DTTI is extracted to generate the Dot Trace Texture Map (DTTM). Second, SIFT feature points are extracted from two InSAR coherent complex images. They are then screened by comparison with DTTI. These selected points are used to calculate the registration relationship, ensuring precise registration of the two images. Third, the interferometric phase diagram is segmented by semantics based on correlation coefficients. For high-quality interferometric phase regions, we use two-dimensional frequency domain rotational symmetric Gaussian Laplace and wavelet domain filters successively. For low-quality regions, we use an average filter to smooth out noise. The superiority of the proposed algorithm was verified through comparative experiments.

An experimental study of the effects of porosity and ratio of fin to channel width on water flow boiling in copper foam fin microchannels

Flow boiling in copper foam fin microchannels (FFMC) has demonstrated better heat transfer performance and flow stability. In this work, systematic experiments are conducted to investigate the effects of structural parameters on flow boiling in five FFMCs with porosities ranging from 0.78 to 0.82 and ratios of fin width to channel width ranging from 0.5 to 2. Five mass fluxes ranging from 68 kg/(m2s) to 408 kg/(m2s) and effective heat fluxes ranging from 2 W/cm2 to 297 W/cm2 are tested. The heat transfer, pressure drop, and flow instability characteristics are analyzed. Generally, the heat transfer in FFMCs follows the “three-zone” model. The heat transfer coefficient first increases with increasing heat flux and then decreases. A “synchronous backflow” phenomenon is observed in FFMCs, which is the direct cause of heat transfer deterioration. Due to the interaction of heat transfer area and the transverse flow resistance through the fins, the heat transfer coefficient increases with decreasing porosity in the low-quality region, while the opposite trend is observed in the high-quality region. The utilization efficiency of the fin area decreases as the fin width increases. An optimal ratio of fin width to channel width exists corresponding to the best heat transfer performance, which equals one in this work. A correlation is developed based on 583 data points considering heat transfer mechanisms in channels and fins. The new correlation predicts the experimental values with a mean absolute percentage error of 9 %, and over 99 % of the data are predicted within a 30 % deviation. A performance evaluation criterion is defined to comprehensively compare the FFMCs, and the FFMC with the porosity of 0.78 demonstrates the best flow boiling performance due to higher heat transfer coefficients and lower pumping powers when the ratio of fin to channel width is equal to one.

Insight into the effect of geographic location and intercropping on contamination characteristics and exposure risk of phthalate esters (PAEs) in tea plantation soils

Phthalate esters (PAEs) are an emerging pollutant due to widespread distribution in environmental mediums that have attracted widespread attention over recent years. However, there is little information about tea plantation soil PAEs. A total of 270 soil samples collected from 45 tea plantations in the major high-quality tea-producing regions of Jiangsu, Zhejiang, and Anhui provinces in China were analyzed for seven PAEs. The detection frequency of PAEs in tea plantation soil was 100%. DBP, DEHP, and DiBP were the main congeners in tea plantation soil. The PAEs concentrations in the upper soil were significantly higher than those in the lower soil. The concentration of tea plantation soil PAEs in Jiangsu Province was significantly lower than those in Zhejiang and Anhui provinces. Intercropping with chestnuts can effectively reduce the contamination level of PAEs in tea plantation soil. Correlation analysis, redundancy analysis, partial correlation analysis, and structural equation modeling methods further confirmed the strong direct influence of factors such as chestnut–tea intercropping, temperature, and agricultural chemicals on the variation of PAEs in tea plantation soil. The health and ecological risk assessments indicated that non-carcinogenic risk was within a safe range and that there was a high carcinogenic risk via the dietary pathway, with DBP posing the highest ecological risk.

A deep learning-based and adaptive region proposal algorithm for semantic segmentation

This paper presents an adaptive and new region proposal algorithm for generating high-quality regions. The main aim of this algorithm is to investigate different features in the proposal generation process. This algorithm is based on bottom-up image segmentation and deep learning techniques. Each primary region is represented by features derived from a convolutional neural network (CNN). The adjacent and similar regions are merged based on a new proposed searching algorithm and a distance function in a hierarchical way. Additionally, various hand-crafted texture features are examined for representing each region. These texture features are not being utilized previously in region proposal generation. This method also applied both texture-based and deep learning-based features, contemporaneously. Furthermore, the proposed region proposal algorithm was evaluated on two significant challenging datasets, including VOC2012 and COCO2017. The resulting proposals show more high-quality regions in high overlaps in comparison to previous region proposal algorithms. More importantly, the results verified that the deep learning-based features together with handcrafted texture features are complement and this fusion can overcome the shortcoming of other approaches. Additionally, the proposed region proposal algorithm is employed in weakly supervised semantic segmentation. The new generated proposals are used to create some new labeled masks. These masks are very useful and effective in the training phase of deep learning. This approach is evaluated on VOC2012 and compared with previous region-based and free region-based methods. The results show the efficacy of the proposed method.

Heat transfer characteristics of R134a flow boiling in a microfin tube under typical ORC pressures based on comparison with a smooth tube

The experimental investigation on the flow boiling heat transfer of an organic working fluid under typical organic Rankine cycle (ORC) operating conditions is insufficient, which directly affects the performance of the ORC system. In this paper, the flow boiling characteristics of R134a in a horizontal microfin tube were experimentally investigated in the reduced pressure range 0.6 ∼ 0.9. The general convection heat transfer characteristics of R134a flow boiling, along with the effects of heat flux, mass flux, pressure, and rewetting conditions, were investigated in a microfin tube and compared with that in a smooth tube. Results show that the main flow regime in the microfin tube under typical ORC operating pressures is the stratified flow with dry-out occurring at the top surface. The microfin effectively enhances the heat transfer and restrains the wall temperature separation through inducing strong helical flow due to buoyancy and centrifugal forces. With the heat flux increasing, the micro-fin strengthens the nucleate boiling by enhancing the vaporization cores, leading to more pronounced heat transfer enhancement compared with that in the smooth tube. The heat transfer on the bottom surface shows almost negligible sensitivity to variations in mass flux. However, in regions of high vapor quality, an increase in mass flux results in higher heat transfer coefficients on the top surface. Due to the interaction between intensified nucleate boiling and weakened gas convection when pressure is increased, it slightly affects the heat transfer coefficient. Unlike the smooth tube, the rewetting in the microfin tube can be maintained within a larger range of vapor quality due to the spiral flow caused by the micro-fin.

Quality-Aware Selective Fusion Network for V-D-T Salient Object Detection.

Depth images and thermal images contain the spatial geometry information and surface temperature information, which can act as complementary information for the RGB modality. However, the quality of the depth and thermal images is often unreliable in some challenging scenarios, which will result in the performance degradation of the two-modal based salient object detection (SOD). Meanwhile, some researchers pay attention to the triple-modal SOD task, namely the visible-depth-thermal (VDT) SOD, where they attempt to explore the complementarity of the RGB image, the depth image, and the thermal image. However, existing triple-modal SOD methods fail to perceive the quality of depth maps and thermal images, which leads to performance degradation when dealing with scenes with low-quality depth and thermal images. Therefore, in this paper, we propose a quality-aware selective fusion network (QSF-Net) to conduct VDT salient object detection, which contains three subnets including the initial feature extraction subnet, the quality-aware region selection subnet, and the region-guided selective fusion subnet. Firstly, except for extracting features, the initial feature extraction subnet can generate a preliminary prediction map from each modality via a shrinkage pyramid architecture, which is equipped with the multi-scale fusion (MSF) module. Then, we design the weakly-supervised quality-aware region selection subnet to generate the quality-aware maps. Concretely, we first find the high-quality and low-quality regions by using the preliminary predictions, which further constitute the pseudo label that can be used to train this subnet. Finally, the region-guided selective fusion subnet purifies the initial features under the guidance of the quality-aware maps, and then fuses the triple-modal features and refines the edge details of prediction maps through the intra-modality and inter-modality attention (IIA) module and the edge refinement (ER) module, respectively. Extensive experiments are performed on VDT-2048 dataset, and the results show that our saliency model consistently outperforms 13 state-of-the-art methods with a large margin. Our code and results are available at https://github.com/Lx-Bao/QSFNet.

Flow condensation inside a multiport mini channel and a rectangular mini channel with pin fin array

In this paper, an experimental study of flow condensation was carried out of R134a in a multiport mini channel and a mini channel with pin fin array. The former comprised 20 parallel rectangular channels with an equivalent diameter of 0.64 mm. The latter is a narrow rectangular mini channel containing 10 rows of diamond pin fins with a staggered configuration, and height and longitudinal/transverse pitch of 0.5 mm and 2 mm respectively. To acquire the local value of heat flux and heat transfer coefficient, air jet impingement cooling devices were adopted to condense the vapor of refrigerants. The effects of vapor quality, mass flux, heat flux, and saturation pressure on flow condensation heat transfer coefficient were investigated with the operating conditions: vapor quality from 1 to 0, mass flux from 160 to 450 kg/(m2s), heat flux from 10.0 to 39.8 kW/m2, and saturation pressure from 600 to 1500 kPa. The experimental results indicated that the pin fin array significantly improves the condensation heat transfer coefficient up to nearly 186 % compared to the multiport mini channel in the high vapor quality region. For both channels, the local heat transfer coefficient increases with an increase in vapor quality, mass flux, and heat flux whereas decreases with increase in saturation pressure. The influence of heat flux and mass flux on heat transfer coefficient was more pronounced in the high vapor quality region than in the low vapor quality region. A sensitivity analysis was performed at different vapor quality levels. The most influential parameters in the smooth channel and the pin fin array are saturation pressure and mass flux, respectively. The relative contribution of heat flux is more obvious in the high vapor quality region. Some of the existing correlations have good prediction performance for the smooth channel experimental data in this paper, and the corresponding MAD is within 20 %. However, their deviations are much greater for the applications of pin fin array.

Localized fabrication of flexible graphene-copper composites via a combined ultrafast laser irradiation and electrodeposition technique

Graphene-metal composites (GMC) possess superior properties, such as high specific strength, and high thermal and electrical conductivity, and have been widely used in many fields. However, the fabrication of GMC still remains to be explored, as existing methods via powder metallurgy and chemical synthesis are difficult to achieve uniform dispersion of graphene, not to mention achieving selective and localized preparation of GMC with complex patterns. In this paper, a simple method for the fabrication of graphene-copper composites (GCC) is firstly proposed by combining flexible laser irradiation and efficient electrodeposition. It not only effectively solves the problems of graphene agglomeration and inhomogeneous dispersion in GCC, but also enables the selective and precise preparation of GCC with spatial micro-patterns. Specifically, an ultrafast picosecond (ps) laser is firstly used to irradiate the polyimide (PI) film to selectively and precisely produce the desired laser-induced graphene (LIG) region. The localized deposition of copper atoms is subsequently made via the electrodeposition step, forming the flexible GCC with a laser-determined spatial pattern. The experimental study of LIG process on PI film has been carried out, and high-quality LIG region is obtained by optimizing the laser scanning path and parameters. On this basis, the regional flexible GCC of 45 μm in thickness with a compact and uniform copper layer of 20 μm has been efficiently fabricated, followed by the detailed characterizations of surface morphology and chemical composition. Furthermore, a significant enhancement of the electrical conductivity has been confirmed, as the measured value is up to 32.8 S/cm at the LIG region and further to about 60.9 S/cm at the GCC area. The mechanical properties of the obtained conductive GCC have been tested and good flexibility is confirmed by nanoindentation and bending tests, verifying its potential for a wide range of applications in flexible electronics, medical and many other fields.

A machine learning framework for neighbor generation in metaheuristic search

This paper presents a methodology for integrating machine learning techniques into metaheuristics for solving combinatorial optimization problems. Namely, we propose a general machine learning framework for neighbor generation in metaheuristic search. We first define an efficient neighborhood structure constructed by applying a transformation to a selected subset of variables from the current solution. Then, the key of the proposed methodology is to generate promising neighbors by selecting a proper subset of variables that contains a descent of the objective in the solution space. To learn a good variable selection strategy, we formulate the problem as a classification task that exploits structural information from the characteristics of the problem and from high-quality solutions. We validate our methodology on two metaheuristic applications: a Tabu Search scheme for solving a Wireless Network Optimization problem and a Large Neighborhood Search heuristic for solving Mixed-Integer Programs. The experimental results show that our approach is able to achieve a satisfactory trade-offs between the exploration of a larger solution space and the exploitation of high-quality solution regions on both applications.

Study on the spatial–temporal evolution and driving mechanism of urban land green use efficiency in the Yellow River Basin cities

Under the current social background of green, sustainable, and high-quality development, countries and regions have increasingly begun to regard green development as a major goal of their economic and social development plans. If the evaluation of urban land use efficiency continues to be based only on economic and social outputs, such research will not conform to the current social development situation. Therefore, this paper introduces negative impacts, such as environmental pollution, into the evaluation system, using the Super-EBM model to measure the urban land green use efficiency (ULGUE) of the Yellow River Basin (YRB) based on panel data from 2011 to 2019. Exploratory spatial data analysis, kernel density estimation, and trend surface analysis were used to study the spatial–temporal evolution characteristics of ULGUE, and the GTWR model was used to explore potential driving mechanisms. The overall ULGUE remained relatively flat from 2011 to 2015, before showing a climbing growth in the post-2015 period; at the sub-basin level, the ULGUE in the midstream area was significantly higher than that in the downstream area, and slightly higher than that in the upstream area, but the rate of ULGUE increase in the upstream area was significantly higher than that in other areas. The spatial aggregation degree of ULGUE showed an increasing trend followed by a decreasing trend, but at the end of the study period, it was still slightly higher than its initial level. The overall ULGUE showed a “polarization–non-differentiation–polarization” trajectory, with the highest efficiency area gradually shifting from the northwest to the central region. Although the driving factors vary between different cities, in general, the roles of technology input and pollution emission factors have gradually increased. This research provides a scientific reference for high-quality development of the YRB and supports the optimal allocation of land resources and regional coordinated development.

An efficient hybrid approach for optimization using simulated annealing and grasshopper algorithm for IoT applications

The multi-objective grasshopper optimization algorithm (MOGOA) is a relatively new algorithm inspired by the collective behavior of grasshoppers, which aims to solve multi-objective optimization problems in IoT applications. In order to enhance its performance and improve global convergence speed, the algorithm integrates simulated annealing (SA). Simulated annealing is a metaheuristic algorithm that is commonly used to improve the search capability of optimization algorithms. In the case of MOGOA, simulated annealing is integrated by employing symmetric perturbation to control the movement of grasshoppers. This helps in effectively balancing exploration and exploitation, leading to better convergence and improved performance.The paper proposes two hybrid algorithms based on MOGOA, which utilize simulated annealing for solving multi-objective optimization problems. One of these hybrid algorithms combines chaotic maps with simulated annealing and MOGOA. The purpose of incorporating simulated annealing and chaotic maps is to address the issue of slow convergence and enhance exploitation by searching high-quality regions identified by MOGOA.Experimental evaluations were conducted on thirteen different benchmark functions to assess the performance of the proposed algorithms. The results demonstrated that the introduction of simulated annealing significantly improved the convergence of MOGOA. Specifically, the IDG (Inverse Distance Generational distance) values for benchmark functions ZDT1, ZDT2, and ZDT3 were smaller than the IDG values obtained by using MOGOA alone, indicating better performance in terms of convergence. Overall, the proposed algorithms exhibit promise in solving multi-objective optimization problems.

Numerical investigation of the gravity effect on two-phase flow and heat transfer of neon condensation inside horizontal tubes

The condensation flow and heat transfer characteristics of cryogenic neon inside horizontal tubes with diameters of 1 mm and 2 mm were numerically investigated. The mass flux ranged from 20 to 187 kg/(m2·s) for gravity levels includes zero gravity, Lunar gravity, Martian gravity, and Earth gravity. The effects of gravity on the condensation flow pattern, local condensate film distribution, and local and global heat transfer were analyzed in detail. The gravity effect on the condensation heat transfer showed three states, including enhancement, deterioration and gravity-independence, which were mainly determined by the mass flux and vapor quality. Under normal gravity, the condensation heat transfer coefficients increased with increasing mass flux and decreasing tube diameter. Under low mass flux, the cryogenic neon underwent enhanced condensation heat transfer under zero gravity and reduced gravity in the high vapor quality region but deteriorated condensation heat transfer in the low vapor quality region. By increasing the mass flux, the gravity effect was effectively suppressed, which resulted in gravity-independent behavior of condensation heat transfer in the high vapor quality region. The gravity effect significantly affected the liquid film distribution in the condensation process, with a more uniform distribution of the liquid film appearing along the circumferential direction as the gravity level decreased. However, due to the low surface tension of the deep cryogenic fluid, the condensation interface was more prone to flow instability under reduced gravity levels.

Compartments of roots and mature leaves are key hubs in the connectivity of tea-plant mycobiomes and are influenced by environmental factors and host age

Understanding the tripartite consortium of crop, mycobiome, and environment is necessary to advance smart farming. Owing to their life cycle of hundreds of years, tea plants are excellent models for studying these entwined relationships; however, observations on this globally important cash crop with numerous health benefits are still rudimentary. Here, the fungal taxa along the soil–tea plant continuum in tea gardens of different ages in famous high-quality tea-growing regions in China were characterized using DNA metabarcoding. Using machine learning, we dissected the spatiotemporal distribution, co-occurrence patterns, assembly, and their associations in different compartments of tea-plant mycobiomes, and further explored how these potential interactions were driven by environmental factors and tree age, and how they influenced the market prices of tea. The results revealed that Compartment niche differentiation was the key driving force behind variation in the tea-plant mycobiome. The mycobiome of roots had the highest specific proportion and convergence and almost did not overlap with the soil. The enrichment ratio of developing leaves to root mycobiome increased with increasing tree age, while mature leaves showed the highest value in the Laobanzhang (LBZ) tea garden with top market prices and displayed the strongest depletion effect on mycobiome association along the soil–tea plant continuum. The balance between determinism and stochasticity in the assembly process was co-driven by compartment niches and life cycle variation. Fungal guild analysis showed that altitude indirectly affected market prices of tea by mediating the abundance of the plant pathogen. The relative importance of plant pathogen and ectomycorrhizae could be used to assess the age of tea. Biomarkers were mainly distributed in soil compartments, and Clavulinopsis miyabeana, Mortierella longata, and Saitozyma sp. may affect the spatiotemporal dynamics of tea-plant mycobiomes and their ecosystem services. Soil properties (mainly total potassium) and tree age indirectly affected the developing leaves via positively influencing the mycobiome of mature leaves. In contrast, the climate directly and significantly drove the mycobiome composition of the developing leaves. Moreover, the proportion of negative correlations in the co-occurrence network positively regulated tea-plant mycobiome assembly, which significantly affected the market prices of tea in the structural equation model with network complexity as hub. These findings indicate that mycobiome signatures play pivotal roles in the adaptive evolution and fungal disease control of tea plants and can help develop better agricultural practices that focus on both plant health and financial profits, and provide a new strategy for assessing tea quality grade and age.

The influences of finite aperture size in photoacoustic computed tomography

In photoacoustic computed tomography (PACT), the “finite aperture effect” is often characterized as a tangential resolution that increases proportionally with the distance from the rotation center. However, this conclusion is based on the inaccurate point-detector assumption used in image reconstruction. In this study, we appropriately modeled the finite size of the acoustic detector in the back-projection (BP) based image reconstruction to improve the accuracy of the time delay calculation and systematically investigated its effects. Our results showed that the main effect of the finite aperture size is the creation of a limited high-quality imaging region (HQIR) around the scanning center, due to the directional sensitivity of the detector. We also demonstrated that the “finite aperture effect” can reduce the optimal number of detectors required for spatial anti-aliasing. These new findings provide novel perspectives for optimizing PACT systems and corresponding reconstruction methods.

Brain tumor classification utilizing deep features derived from high-quality regions in MRI images

The accurate and rapid detection of brain tumors is crucial for expediting patient rehabilitation and saving lives. Brain tumors exhibit considerable variation in size, shape, and appearance, necessitating patient-specific treatment options. Radiologists are required to possess a high level of experience and skill to manually diagnose brain cancers. However, manual tumor detection can be inefficient, error-prone, and time-consuming. Developing a practical solution to overcome these limitations is essential. This work presents a less labor-intensive approach to diagnosing brain cancers using MRI imaging. The proposed method enhances an image's visual quality by employing a low-complexity algorithm. Prior to segmentation, morphological analysis is conducted to exclude non-tumor regions from the images. Employing segmentation and clustering techniques, high-quality tumor regions are identified. Based on their rankings, multiple deep neural networks extract features from these regions. An adaptive fusion network creates a hybrid feature vector, which is then used in conjunction with multi-class SVM to classify tumors. By supplementing the training data, overfitting is minimized. The presented system is trained and evaluated using a publicly accessible brain tumor dataset. In comparison to previous approaches, the proposed solution enhances the resilience and automation of the entire diagnostic process, enabling the healthcare industry to achieve an overall accuracy of approximately 98.98%.

Identification of a secondary Q-marker in high-quality ecotypes of Carthamus tinctorius L. and exploration of the target preference.

Safflower (Carthamus tinctorius) has the efficacy for promoting blood circulation and preventing cardiovascular and Alzheimer's diseases and is thus a valuable medicinal and functional food plant. However, how to evaluate high-quality safflower is still a problem. To differentiate intraspecies ecotypes and illustrate the mechanisms of differential metabolites of C. tinctorius from different regions, this study combined the widely targeted metabolome, weighted network pharmacology, and molecular docking to filter bioactive compounds and predict the target preference. The results indicated that kaempferol is suitable as a secondary Q-marker to differentiate intraspecies ecotypes. In secondary metabolites, the average content of kaempferol and its derivates in C. tinctorius from Sichuan is three times that of other areas, which have the potential for the targeted medicine of CA2 and TNF. In volatile metabolites, isoaromadendrene epoxide has the potential as a specifically targeted medicine of RXRA. The change of the target preference could be the reason for the difference in drug efficacy among different varieties of C. tinctorius. It is reasonable that Sichuan was recognized as a high-quality ecotype producing region of C. tinctorius in China, which promotes blood circulation and removes blood stasis. This study provides an innovative method to differentiate intraspecies ecotypes and explore their target preference.

Temporal Speciation Network for Few-Shot Object Detection

Recently, few-shot object detection (FSOD) has become an increasing research focus, which can largely alleviate the heavy dependency on expensive annotations in the traditional object detection task. However, existing FSOD approaches fail to generate sufficient high-quality positive region proposals which are the key to detection performance, due to the lack of informative knowledge from base classes and non-specific alteration for novel classes. To address the problem, this paper presents a simple yet effective few-shot object detection framework referred to as Temporal Speciation Network (TeSNet) with an evolving training, which improves the diversity and rationality of positive proposal generation. Our TeSNet, imitating the natural evolution which relies on inheritation and mutation, correspondingly consists of two key components: a Selective Recombination Module (SRM) for effectively inheriting from base classes and a Mutational Region Proposal Network (MRPN) for flexibly mutating according to the unique traits of novel samples. Specifically, SRM selects and reorganizes relevant base categories, and further instantiates diverse individuals to ensure the diversity of positive proposals. MRPN adapts the parameters trained on base classes aiming for accurately locating positive proposals. Extensive experiments are conducted on several commonly-used datasets, in which our TeSNet achieves state-of-the-art results and outperforms baselines by large margin.