Efficient click-based interactive segmentation for medical image with improved Plain-ViT.

The primary objective of interactive medical image segmentation systems is to achieve more precise segmentation outcomes with reduced human intervention. This endeavor holds significant clinical importance for both pre-diagnostic pathological assessments and prognostic recovery. Among the various interaction methods available, click-based interactions stand out as an intuitive and straightforward approach compared to alternatives such as graffiti, bounding boxes, and extreme points. To improve the model's ability to interpret click-based interactions, we propose a comprehensive interactive segmentation framework that leverages an iterative weighted loss function based on user clicks. To enhance the segmentation capabilities of the Plain-ViT backbone, we introduce a Residual Multi-Headed Self-Attention encoder with hierarchical inputs and residual connections, offering multiple perspectives on the data. This innovative architecture leads to a remarkable improvement in segmentation model performance. In this research paper, we assess the robustness of our proposed framework using a self-compiled T2-MRI image dataset of the prostate and three publicly available datasets containing images of other organs. Our experimental results convincingly demonstrate that our segmentation model surpasses existing state-of-the-art methods. Furthermore, the incorporation of an iterative loss function training strategy significantly accelerates the model's convergence rate during interactions. In the prostate dataset, we achieved an impressive Intersection over Union (IoU) score of 88.11% and Number of Clicks(NoC) at 80% are 7.03 clicks.

Predicting ambient PM2.5 concentrations via time series models in Anhui Province, China.

Due to rapid expansion in the global economy and industrialization, PM2.5 (particles smaller than 2.5 µm in aerodynamic diameter) pollution has become a key environmental issue. The public health and social development directly affected by high PM2.5 levels. In this paper, ambient PM2.5 concentrations along with meteorological data are forecasted using time series models, including random forest (RF), prophet forecasting model (PFM), and autoregressive integrated moving average (ARIMA) in Anhui province, China. The results indicate that the RF model outperformed the PFM and ARIMA in the prediction of PM2.5 concentrations, with cross-validation coefficients of determination R2, RMSE, and MAE values of 0.83, 10.39 µg/m3, and 6.83 µg/m3, respectively. PFM achieved the average results (R2 = 0.71, RMSE = 13.90 µg/m3, and MAE = 9.05 µg/m3), while the predicted results by ARIMA are comparatively poorer (R2 = 0.64, RMSE = 15.85 µg/m3, and MAE = 10.59 µg/m3) than RF and PFM. These findings reveal that the RF model is the most effective method for predicting PM2.5 and can be applied to other regions for new findings.

Global production patterns: Understanding the relationship between greenhouse gas emissions, agriculture greening and climate variability.

Climate change due to increased greenhouse gas emissions (GHG) in the atmosphere has been consistently observed since the mid-20th century. The profound influence of global climate change on greenhouse gas (GHG) emissions, encompassing carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O), has established a vital feedback loop that contributes to further climate change. This intricate relationship necessitates a comprehensive understanding of the underlying feedback mechanisms. By examining the interactions between global climate change, soil, and GHG emissions, we can elucidate the complexities of CO2, CH4, and N2O dynamics and their implications. In this study, we evaluate the global climate change relationship with GHG globally in 246 countries. We find a robust positive association between climate and GHG emissions. By 2100, GHG emissions will increase in all G7 countries and China while decreasing in the United Kingdom based on current economic growth policies, resulting in a net global increase, suggesting that climate-driven increase in GHG and climate variations impact crop production loss due to soil impacts and not provide climate adaptation. The study highlights the diverse strategies employed by G7 countries in reducing GHG emissions, with France leveraging nuclear power, Germany focusing on renewables, and Italy targeting its industrial and transportation sectors. The UK and Japan are making significant progress in emission reduction through renewable energy, while the US and Canada face challenges due to their industrial activities and reliance on fossil fuels.

Interpretable Machine Learning for Personalized Medical Recommendations: A LIME-Based Approach.

Chronic diseases are increasingly major threats to older persons, seriously affecting their physical health and well-being. Hospitals have accumulated a wealth of health-related data, including patients' test reports, treatment histories, and diagnostic records, to better understand patients' health, safety, and disease progression. Extracting relevant information from this data enables physicians to provide personalized patient-treatment recommendations. While collaborative filtering techniques and classical algorithms such as naive Bayes, logistic regression, and decision trees have had notable success in health-recommendation systems, most current systems primarily inform users of their likely preferences without providing explanations. This paper proposes an approach of deep learning with a local interpretable model-agnostic explanations (LIME)-based interpretable recommendation system to solve this problem. Specifically, we apply the proposed approach to two chronic diseases common in older adults: heart disease and diabetes. After data preprocessing, we use six deep-learning algorithms to form interpretations. In the heart-disease data set, the actual model recommendation of multi-layer perceptron and gradient-boosting algorithm differs from the local model's recommendation of LIME, which can be used as its approximate prediction. From the feature importance of these two algorithms, it can be seen that the CholCheck, GenHith, and HighBP features are the most important for predicting heart disease. In the diabetes data set, the actual model predictions of the multi-layer perceptron and logistic-regression algorithm were little different from the local model's prediction of LIME, which can be used as its approximate recommendation. Moreover, from the feature importance of the two algorithms, it can be seen that the three features of glucose, BMI, and age were the most important for predicting heart disease. Next, LIME is used to determine the importance of each feature that affected the results of the calculated model. Subsequently, we present the contribution coefficients of these features to the final recommendation. By analyzing the impact of different patient characteristics on the recommendations, our proposed system elucidates the underlying reasons behind these recommendations and enhances patient trust. This approach has important implications for medical recommendation systems and encourages informed decision-making in healthcare.

Deep reinforcement learning enables adaptive-image augmentation for automated optical inspection of plant rust.

This study proposes an adaptive image augmentation scheme using deep reinforcement learning (DRL) to improve the performance of a deep learning-based automated optical inspection system. The study addresses the challenge of inconsistency in the performance of single image augmentation methods. It introduces a DRL algorithm, DQN, to select the most suitable augmentation method for each image. The proposed approach extracts geometric and pixel indicators to form states, and uses DeepLab-v3+ model to verify the augmented images and generate rewards. Image augmentation methods are treated as actions, and the DQN algorithm selects the best methods based on the images and segmentation model. The study demonstrates that the proposed framework outperforms any single image augmentation method and achieves better segmentation performance than other semantic segmentation models. The framework has practical implications for developing more accurate and robust automated optical inspection systems, critical for ensuring product quality in various industries. Future research can explore the generalizability and scalability of the proposed framework to other domains and applications. The code for this application is uploaded at https://github.com/lynnkobe/Adaptive-Image-Augmentation.git.

Safe Decision Controller for Autonomous DrivingBased on Deep Reinforcement Learning inNondeterministic Environment.

Autonomous driving systems are crucial complicated cyber-physical systems that combine physical environment awareness with cognitive computing. Deep reinforcement learning is currently commonly used in the decision-making of such systems. However, black-box-based deep reinforcement learning systems do not guarantee system safety and the interpretability of the reward-function settings in the face of complex environments and the influence of uncontrolled uncertainties. Therefore, a formal security reinforcement learning method is proposed. First, we propose an environmental modeling approach based on the influence of nondeterministic environmental factors, which enables the precise quantification of environmental issues. Second, we use the environment model to formalize the reward machine's structure, which is used to guide the reward-function setting in reinforcement learning. Third, we generate a control barrier function to ensure a safer state behavior policy for reinforcement learning. Finally, we verify the method's effectiveness in intelligent driving using overtaking and lane-changing scenarios.

Assessing the ambient air quality patterns associated to the COVID-19 outbreak in the Yangtze River Delta: A random forest approach.

The novel coronavirus (COVID-19), first identified at the end of December 2019, has significant impacts on all aspects of human society. In this study, we aimed to assess the ambient air quality patterns associated to the COVID-19 outbreak in the Yangtze River Delta (YRD) region using a random forest (RF) model. To estimate the accuracy of the model, the cross-validation (CV), determination coefficient R2, root mean squared error (RMSE) and mean absolute error (MAE) were used. The results demonstrate that the RF model achieved the best performance in the prediction of PM10 (R2 = 0.78, RMSE = 8.81 µg/m3), PM2.5 (R2 = 0.76, RMSE = 6.16 µg/m3), SO2 (R2 = 0.76, RMSE = 0.70 µg/m3), NO2 (R2 = 0.75, RMSE = 4.25 µg/m3), CO (R2 = 0.81, RMSE = 0.4 µg/m3) and O3 (R2 = 0.79, RMSE = 6.24 µg/m3) concentrations in the YRD region. Compared with the prior two years (2018-19), significant reductions were recorded in air pollutants, such as SO2 (-36.37%), followed by PM10 (-33.95%), PM2.5 (-32.86%), NO2 (-32.65%) and CO (-20.48%), while an increase in O3 was observed (6.70%) during the COVID-19 period (first phase). Moreover, the YRD experienced rising trends in the concentrations of PM10, PM2.5, NO2 and CO, while SO2 and O3 levels decreased in 2021-22 (second phase). These findings provide credible outcomes and encourage the efforts to mitigate air pollution problems in the future.

AI-based object detection latest trends in remote sensing, multimedia and agriculture applications.

Object detection is a vital research direction in machine vision and deep learning. The object detection technique based on deep understanding has achieved tremendous progress in feature extraction, image representation, classification, and recognition in recent years, due to this rapid growth of deep learning theory and technology. Scholars have proposed a series of methods for the object detection algorithm as well as improvements in data processing, network structure, loss function, and so on. In this paper, we introduce the characteristics of standard datasets and critical parameters of performance index evaluation, as well as the network structure and implementation methods of two-stage, single-stage, and other improved algorithms that are compared and analyzed. The latest improvement ideas of typical object detection algorithms based on deep learning are discussed and reached, from data enhancement, a priori box selection, network model construction, prediction box selection, and loss calculation. Finally, combined with the existing challenges, the future research direction of typical object detection algorithms is surveyed.

Evaluating the impact of roads on the diversity pattern and density of trees to improve the conservation of species.

Roadside trees alter biotic and abiotic factors of plants diversity in an ecosystem. Rows of plants grow along the roadside due to the interplay between the arrival of propagule and seedling establishment, which depends on the road's specifications, land pattern, and road administration and protection practices. A field study was conducted to measure the roadside tree diversity in the city of Karachi (Pakistan). A total of 180 plots, divided into three primary road groups, were surveyed. The highest quantity of tree biomass per unit area was found on wide roads, followed by medium roads. On narrow roads, the least biomass was detected. A single species or a limited number of species dominated the tree community. Conocarpus erectus was the most dominant non-native species on all types of sidewalks or roadsides, followed by Guaiacum officinale. A total of 76 species (32 non-natives and 44 natives) that were selectively spread along the roadsides of the city were studied. There was a significant difference in phylogenetic diversity (PD), phylogenetic mean pairwise distance (MPD), and phylogenetic mean nearest taxon distance (MNTD) among wide, medium, and narrow roads. Management practices have a significant positive correlation with diversity indices. Our study identified patterns of diversity in roadside trees in Karachi. It provides the basis for future planning for plant protection, such as the protection of plant species, the maintenance of plant habitats, and the coordination of plant management in Karachi.

Assessing the change of ambient air quality patterns in Jiangsu Province of China pre-to post-COVID-19.

Following the outbreak of the novel coronavirus in early 2020, to effectively prevent the spread of the disease, major cities across China suspended work and production. While the rest of the world struggles to control COVID-19, China has managed to control the pandemic rapidly and effectively with strong lockdown policies. This study investigates the change in air pollution (focusing on the air quality index (AQI), six ambient air pollutants nitrogen dioxide (NO2), ozone (O3), sulphur dioxide (SO2), carbon monoxide (CO), particulate matter with aerodynamic diameters ≤10 µm (PM10) and ≤2.5 µm (PM2.5)) patterns for three periods: pre-COVID (from 1 January to May 30, 2019), active COVID (from 1 January to May 30, 2020) and post-COVID (from 1 January to May 30, 2021) in the Jiangsu province of China. Our findings reveal that the change in air pollution from pre-COVID to active COVID was greater than in previous years due to the government's lockdown policies. Post-COVID, air pollutant concentration is increasing. Mean change PM2.5 from pre-COVID to active COVID decreased by 18%; post-COVID it has only decreased by 2%. PM10 decreased by 19% from pre-COVID to active COVID, but post-COVID pollutant concentration has seen a 23% increase. Air pollutants show a positive correlation with COVID-19 cases among which PM2.5, PM10 and NO2 show a strong correlation during active COVID-19 cases. Metrological factors such as minimum temperature, average temperature and humidity show a positive correlation with COVID-19 cases while maximum temperature, wind speed and air pressure show no strong positive correlation. Although the COVID-19 pandemic had numerous negative effects on human health and the global economy, the reduction in air pollution and significant improvement in ambient air quality likely had substantial short-term health benefits; the government must implement policies to control post-COVID environmental issues.

A hybrid approach to forecast the COVID-19 epidemic trend.

Studying the progress and trend of the novel coronavirus pneumonia (COVID-19) transmission mode will help effectively curb its spread. Some commonly used infectious disease prediction models are introduced. The hybrid model is proposed, which overcomes the disadvantages of the logistic model's inability to predict the number of confirmed diagnoses and the drawbacks of too many tuning parameters of the SEIR (Susceptible, Exposed, Infectious, Recovered) model. The realization and superiority of the prediction of the proposed model are proven through experiments. At the same time, the influence of different initial values of the parameters that need to be debugged on the hybrid model is further studied, and the mean error is used to quantify the prediction effect. By forecasting epidemic size and peak time and simulating the effects of public health interventions, this paper aims to clarify the transmission dynamics of COVID-19 and recommend operation suggestions to slow down the epidemic. It is suggested that the quick detection of cases, sufficient implementation of quarantine and public self-protection behaviours are critical to slow down the epidemic.

Advance hybrid medical watermarking algorithm using speeded up robust features and discrete cosine transform.

In the continuous development of computer network technology, multimedia technology and information technology, digitization has become the main means of displaying information, thus facilitating the storage, copying and dissemination of digital multimedia information. In this context, there are no restrictions on arbitrary editing, copying, modification, and dissemination of digital images, music, etc., which leads to various social problems such as information security, copyright disputes, and piracy. With the advancement of networks and multimedia, digital watermarking technology has received worldwide attention as an effective method of copyright protection. Improving the anti-geometric attack ability of digital watermarking algorithms using image feature-based algorithms have received extensive attention. This paper proposes a novel robust watermarking algorithm based on SURF-DCT perceptual hashing (Speeded Up Robust Features and Discrete Cosine Transform), namely blind watermarking. The algorithm firstly uses the affine transformation with a feature matrix and chaotic encryption technology to preprocess the watermark image, enhance the confidentiality of the watermark, and perform block and DCT coefficients extraction on the carrier image, and then uses the positive and negative quantization rules to modify the DCT coefficients. The embedding of the watermark is completed, and the blind extraction of the watermark realized. Experiments show that the algorithm has good invisibility and strong robustness against conventional and geometric attacks and can effectively protect the security of images with NC value more than 90%.

A Clinical Decision Support Framework for Heterogeneous Data Sources.

To keep pace with the developments in medical informatics, health medical data is being collected continually. But, owing to the diversity of its categories and sources, medical data has become so complicated in many hospitals that it now needs a clinical decision support (CDS) system for its management. To effectively utilize the accumulating health data, we propose a CDS framework that can integrate heterogeneous health data from different sources such as laboratory test results, basic information of patients, and health records into a consolidated representation of features of all patients. Using the electronic health medical data so created, multilabel classification was employed to recommend a list of diseases and thus assist physicians in diagnosing or treating their patients' health issues more efficiently. Once the physician diagnoses the disease of a patient, the next step is to consider the likely complications of that disease, which can lead to more diseases. Previous studies reveal that correlations do exist among some diseases. Considering these correlations, a k-nearest neighbors algorithm is improved for multilabel learning by using correlations among labels (CML-kNN). The CML- kNN algorithm first exploits the dependence between every two labels to update the origin label matrix and then performs multilabel learning to estimate the probabilities of labels by using the integrated features. Finally, it recommends the top N diseases to the physicians. Experimental results on real health medical data establish the effectiveness and practicability of the proposed CDS framework.

Recommendation system for immunization coverage and monitoring.

Immunization averts an expected 2 to 3 million deaths every year from diphtheria, tetanus, pertussis (whooping cough), and measles; however, an additional 1.5 million deaths could be avoided if vaccination coverage was improved worldwide. 11 Data source for immunization records of 1.5 M: http://www.who.int/mediacentre/factsheets/fs378/en/ New vaccination technologies provide earlier diagnoses, personalized treatments and a wide range of other benefits for both patients and health care professionals. Childhood diseases that were commonplace less than a generation ago have become rare because of vaccines. However, 100% vaccination coverage is still the target to avoid further mortality. Governments have launched special campaigns to create an awareness of vaccination. In this paper, we have focused on data mining algorithms for big data using a collaborative approach for vaccination datasets to resolve problems with planning vaccinations in children, stocking vaccines, and tracking and monitoring non-vaccinated children appropriately. Geographical mapping of vaccination records helps to tackle red zone areas, where vaccination rates are poor, while green zone areas, where vaccination rates are good, can be monitored to enable health care staff to plan the administration of vaccines. Our recommendation algorithm assists in these processes by using deep data mining and by accessing records of other hospitals to highlight locations with lower rates of vaccination. The overall performance of the model is good. The model has been implemented in hospitals to control vaccination across the coverage area.