Search

Population based health data collection and analysis are important in epidemiological research. In recent years, with the rapid development of big data, Internet and cloud computing, artificial intelligence has gradually attracted attention of epidemiological researchers. More and more researchers are trying to use artificial intelligence algorithms for genome sequencing and medical image data mining, and for disease diagnosis, risk prediction and others. In recent years, machine learning, a branch of artificial intelligence, has been widely used in epidemiological research. This paper summarizes the key fields and progress in the application of machine learning in epidemiology, reviews the development history of machine learning, analyzes the classic cases and current challenges in its application in epidemiological research, and introduces the current application scenarios and future development trends of machine learning and artificial intelligence algorithms for the better exploration of the epidemiological research value of massive medical health data in China.

In the study of age estimation in living individuals, a lot of data needs to be analyzed by mathematical statistics, and reasonable medical statistical methods play an important role in data design and analysis. The selection of accurate and appropriate statistical methods is one of the key factors affecting the quality of research results. This paper reviews the principles and applicable principles of the commonly used medical statistical methods such as descriptive statistics, difference analysis, consistency test and multivariate statistical analysis, as well as machine learning methods such as shallow learning and deep learning in the age estimation research of living individuals, and summarizes the relevance and application prospects between medical statistical methods and machine learning methods. This paper aims to provide technical guidance for the age estimation research of living individuals to obtain more scientific and accurate results.

Wound age estimation is the core content in the practice of forensic medicine. Accurate estimation of wound age is a scientific question that needs to be urgently solved by forensic scientists at home and abroad. Metabolomics techniques can effectively detect endogenous metabolites produced by internal or external stimulating factors and describe the dynamic changes of metabolites in vivo . It has the advantages of strong operability, high detection efficiency and accurate quantitative results. Machine learning algorithm has special advantages in processing high-dimensional data sets, which can effectively mine biological information and truly reflect the physiological, disease or injury state of the body. It is a new technical means for efficiently processing high-throughput big data. This paper reviews the status and advantages of metabolomic techniques combined with machine learning algorithm in the research of wound age estimation, and provides new ideas for this research.

Land subsidence is a worldwide geological hazard. Differential land subsidence has posed the major threat to urban infrastructure, linear rail transit and underground space development and utilization, and also restricted the sustainable development of the economy and society. This paper systematically elaborates on the research progresses on land deformation acquisition, evolution mechanism and simulation of land subsidence, and focuses on the land deformation acquisition technology based on InSAR monitoring and multi-source deformation data fusion, as well as the correlation analysis, statistical analysis, machine learning and other methods to analyze the relationship between the evolution of land subsidence and various influencing factors based on geotechnical experiment and long time series observation data. On this basis, the advantages and disadvantages of land subsidence simulation models such as groundwater flow field-land deformation model, mathematical statistical model and machine learning model are explored. It is found that multi-source deformation data fusion can improve the spatiotemporal resolution of land deformation. The differences in geological structure, lithology, groundwater exploitation, and dynamic and static loads are factors contributing to the differential evolution of land subsidence. The difficulty in balancing the computational efficiency and interpretability of mathematical models for land subsidence is the main problem in simulation. According to the literature review, the current researches mainly focus on land subsidence caused by groundwater over-exploitation. This paper further proposes the future research directions for land subsidence, under the background of climate change, new hydrological condition and dataset, and based on the fusion of data through remote sensing and field observations, integrating the latest progress of InSAR, GeoAI, cloud platform and other technologies to reveal the evolution mechanism of land subsidence considering the climate change and anthropic activities and provide technical support for regional land subsidence prevention and urban safety.

The unique flavor properties and rich nutrients of traditional fermented food are closely related to its complex and variable microbial structure, which also makes it difficult to control the quality of final fermented product. In order to explore the changes of microbial structure and sensory property and nutritional property in the process of food fermentation and the internal relationship between them, the data analysis process is a key step. Therefore, it is necessary to establish a fast and accurate data analysis method for quality control of fermented food. Machine learning has the advantages of high-dimensional simplification rate, large data throughput and high prediction accuracy, showing great application potential in the field of quality control of fermented food. Hence, machine learning has become one of the research hotspots. This paper reviews the application of machine learning in the quality control of fermented food. On the basis of an overview of common models of machine learning, this paper systematically summarizes the application of machine learning in the prediction of microbial structure evolution, flavor compound composition analysis and customization of personalized consumption in the process of food fermentation. The problems and developmental trends in the application of machine learning to quality control of traditional fermented food are summarized and prospected. Although the application of machine learning in fermented food is still confined by the problems such as insufficient general applicability of the model, limited quality indicators, and limited personalized consumption scenario, etc., with the iterative update of the technical model, the adaptation for multi-factors and whole process, and the application expansion in the background of personalized consumption, machine learning will show a greater value for practical application in the field of fermented food. The purpose of this study is to provide guidance for the further application of machine learning in the standardized and controllable production of traditional fermented food.

Photoplethysmography (PPG) is often affected by interference, which could lead to incorrect judgment of physiological information. Therefore, performing a quality assessment before extracting physiological information is crucial. This paper proposed a new PPG signal quality assessment by fusing multi-class features with multi-scale series information to address the problems of traditional machine learning methods with low accuracy and deep learning methods requiring a large number of samples for training. The multi-class features were extracted to reduce the dependence on the number of samples, and the multi-scale series information was extracted by a multi-scale convolutional neural network and bidirectional long short-term memory to improve the accuracy. The proposed method obtained the highest accuracy of 94.21%. It showed the best performance in all sensitivity, specificity, precision, and F1-score metrics, compared with 6 quality assessment methods on 14 700 samples from 7 experiments. This paper provides a new method for quality assessment in small samples of PPG signals and quality information mining, which is expected to be used for accurate extraction and monitoring of clinical and daily PPG physiological information.

Multi-criteria priority ranking involves determining priorities through calculations for a multitude of tasks as per potentially conflicting criteria, to aid decision-makers in efficient allocation of resources and prioritization of tasks based on their criticality. This process is essential for optimizing resource allocation and improving overall work efficiency. Over the past two decades, the field of priority ranking has seen the development of numerous mature classic methods through continuous modifications and expansions. Furthermore, new sorting methods rooted in advancing technologies in the era of big data have also emerged. To assist decision-makers in evaluating and selecting the most suitable multi-criteria prioritization methods for priority ranking needs in specific application scenarios, this paper first elucidates differences across various multi-criteria prioritization methods regarding criteria substitutability, advantages and disadvantages, data input/output requirements, and applicable scenes. It further categorizes existing multi-criteria task prioritization methods into four main types: full aggregation methods, inferiority ranking methods, reference level methods, and data processing-based prioritization methods. Additionally, it discusses in detail the basic principles, implementation steps, latest research findings, and pros and cons of representative methods within each category. Finally, this paper summarizes current research hotspots, including the analytic hierarchy process, fuzzy theory, and AI-assisted ranking technologies, while suggesting future research pathways covering AI-assisted big data ranking optimization, ranking stability enhancement, and intelligent preference identification.

In the study of age estimation in living individuals, a lot of data needs to be analyzed by mathematical statistics, and reasonable medical statistical methods play an important role in data design and analysis. The selection of accurate and appropriate statistical methods is one of the key factors affecting the quality of research results. This paper reviews the principles and applicable principles of the commonly used medical statistical methods such as descriptive statistics, difference analysis, consistency test and multivariate statistical analysis, as well as machine learning methods such as shallow learning and deep learning in the age estimation research of living individuals, and summarizes the relevance and application prospects between medical statistical methods and machine learning methods. This paper aims to provide technical guidance for the age estimation research of living individuals to obtain more scientific and accurate results.

The safety in the coal-producing mines in China is continuously improving, but coal and gas outburst accidents still occur. The prediction of coal and gas outbursts allows the scientific application of outburst prevention measures, which can ensure the safe coal mining to a certain extent. Machine learning is an interdisciplinary field involving probability theory, statistics, and computer science, which can explore the nonlinear relationship between outburst accidents and its associated indicators. The application of machine learning in coal and gas outburst prediction has received relatively widespread attention, and with the rapid progress of artificial intelligence and computer technology, it will play a greater role in the field of outburst prediction. Therefore, this paper provides a comprehensive review of the research on machine learning in coal and gas outburst prediction, analyzes the difficulties in outburst prediction and prospects its development direction. Firstly, the paper provides a brief overview of the research status on the hypothesis, occurrence mechanism, and prediction index selection of coal and gas outbursts. Then, it summarizes the research progress in the field of outburst prediction, including the application of support vector machines, neural networks, extreme learning machines, and ensemble learning algorithms. In addition, it also points out the existing problems in the current research, such as imbalanced samples, missing data indicators, and small sample sizes. Finally, the paper gives an outlook on the developments of coal and gas outburst prediction based on machine learning, including improving algorithm performance, optimizing feature engineering, and increasing sample size. With the continuous improvement of computer performance, more powerful models may be proposed, which can further improve the prediction accuracy of outburst accidents.

The North Mountain landslide in Lijie Town has been in a long-term creeping deformation state and has experienced multiple landslide and debris flow disasters. Monitoring the surface deformation of landslide to grasp the surface deformation pattern of disaster body is a reliable basis for realizing early warning prediction of geological disaster. In this paper, a machine learning model is introduced to predict the relevant data, and a long and short-term memory network is used to predict the landslide deformation by monitoring the displacement data of North Mountain in Lijie, and the prediction results are compared with the actual data and analyzed. In this paper, root mean square error , mean absolute error , coefficient of determination and explainable variance are used to evaluate the prediction results, among which the coefficient of determination and explainable variance reach 0.99. It shows that the long short-term memory network used in this paper achieves good prediction performance in the prediction of landslide deformation in the North Mountain of Lijie.

Currently, algorithms such as support vector machine (SVM) and random forest (RF) are less applied in the field of coal mine roadway support. The paper studies the applicability of different machine learning models for support parameter design.Thus a higher performance model would be established to achieve reasonable and scientific design of anchor bolt support. Firstly, it is suggested to establish an intelligent prediction database for coal mine roadway support. Through on-site research, questionnaire survey, and literature search, the coal mine roadway samples are collected. The data is processed using methods such as filling in missing values, modifying outliers in box charts, and removing local abnormal factors to establish a coal roadway support database. The paper proposes a coal roadway support parameter prediction model based on synthetic minority oversampling technique (SMOTE) - genetic algorithm (GA) - SVM. The data in the database is divided into training and testing sets. The SMOTE technology is used to balance training samples, and improve the model's fitting capability for minority class samples. The training process uses GA to globally optimize the hyperparameters of SVM, further improving the overall performance of the model. The test results show that the classificaton precision of the SMOTE-GA-SVM model reaches 83.8%, which is 21.8% higher than the traditional SVM model. The machine learning methods such as SVM, artificial neural network (ANN), RF, AdaBoost (ADA), and naive Bayesian classifier (NBC) are introduced into the prediction of coal roadway anchor support parameters. The corresponding support parameter prediction models are established. The comparison results showed that the best to worst prediction models are ranked as SMOTE-GA-SVM, RF, GA-ANN, SVM, NBC, and ADA, with an average classificaton precision of 69.9%. The result verifies the feasibility of machine learning methods in predicting the parameters of coal roadway bolt support. The SMOTE-GA-SVM model is applied in Shanxi Huobaoganhe Coal Mine Co., Ltd., with a precision of 87.5% and strong applicability and reliability.

As an important branch of artificial intelligence, machine learning is widely used in various fields. Machine learning has similarity to classical statistical methods, but can solve many problems which are difficult for traditional statistics, so it is one of the important tools in epidemiological research. This paper introduced 9 common algorithms of machine learning and summarized their characteristics and applications in epidemiological research. Readers could choose appropriate machine learning method according to the research purpose for the better application of machine learning in epidemiological research.

The performance of machine learning-based radar target recognition models is determined by the respective model and data to be analyzed. Currently, radar target recognition performance evaluation is based on accuracy metrics, but this method does not include the evaluation metrics regarding the impact of data quality on recognition performance. Data separability describes the degree of mixture of samples from different categories. Furthermore, the data separability metric is independent of the model recognition process. By incorporating it into the recognition evaluation process, recognition difficulty can be quantified, and a benchmark for recognition results can be provided in advance. Therefore, in this paper, we propose a data separability metric based on the rate-distortion theory. Extensive experiments on multiple simulated datasets demonstrated that the proposed metric can compare the separability of multivariate Gaussian datasets. Furthermore, by combining it with the Gaussian mixture model, the designed metric method could overcome the limitation of the rate-distortion function, capture the data’s local separable characteristics, and improve the evaluation accuracy of the overall data separability. Subsequently, we applied the proposed metric to evaluate the recognition difficulty in real datasets, the results of which validated its strong correlation with average recognition accuracy. In the experiments on evaluating the effectiveness of convolutional neural network modules, we first quantified and analyzed the separability trend of the feature extracted by each module during the testing phase. Further, we incorporated the proposed metric as a feature separability loss during the training phase to participate in the network optimization process, guiding the network to extract a more separable feature. This paper provides a new perspective for evaluating and improving the neural network recognition performance in terms of feature separability.

Machine learning is an important aspect of modern computer technology, and the support vector machine method has been widely used in all walks of life because of its good performance in recent years. The performance of support vector machine can be measured by VC (Vapnik-Chervonenkis) dimension, which is an index to measure the complexity of the machine. In theory, low VC dimension can be well generalized. However, for some classifier methods based on traditional support vector machine, the upper bound of VC dimension of support vector machine may be infinite when dealing with a variety of data. Although good results have been obtained in practice and application, it can not guarantee good generalization, resulting in poor results for some special data. Therefore, this paper proposes an improved LSSVM (least squares support vector machines) algorithm. Based on LSSVM algorithm, this paper minimizes the upper bound of VC dimension and finds the desired best projection scheme. Finally, this paper brings it into LSSVM algorithm to classify the data. Experimental results show that the error rate of the classifier is lower than that of the traditional least squares support vector machine on the benchmark dataset, which means that the proposed algorithm makes the test accuracy better than the comparison algorithm with the approximate number of support vectors, and improves the generalization ability of the algorithm.

Objective: To screen the influencing factors of hypertensive heart disease (HHD), establish the predictive model of HHD, and provide early warning for the occurrence of HHD. Methods: Select the patients diagnosed as hypertensive heart disease or hypertensionfrom January 1, 2016 to December 31, 2019, in the medical data science academy of a medical school. Influencing factors were screened through single factor and multi-factor analysis, and R software was used to construct the logistics model, random forest (RF) model and extreme gradient boosting (XGBoost) model. Results: Univariate analysis screened 60 difference indicators, and multifactor analysis screened 18 difference indicators (P<0.05). The area under the curve (AUC) of Logistics model, RF model and XGBoost model are 0.979, 0.983 and 0.990, respectively. Conclusion: The results of the three HHD prediction models established in this paper are stable, and the XGBoost prediction model has a good diagnostic effect on the occurrence of HHD.

In recent years,machine learning has not only achieved remarkable results in conventional fields such as computer vision and natural language processing,but also been widely applied to process sensitive data such as face images,financial data and medical information.Recently,researchers find that machine learning models will remember the data in their training sets,making them vulnerable to membership inference attacks,that is,the attacker can infer whether the given data exists in the training set of a specific machine learning model.The success of membership inference attacks may lead to serious individual privacy leakage.For example,the existence of a patient's medical record in a hospital's analytical training set reveals that the patient was once a patient there.The paper first introduces the basic principle of membership inference attacks,and then systematically summarizes and classifies the representative research achievements on membership inference attacks and defenses in recent years.In particular,how to attack and defend under different conditions is described in detail.Finally,by reviewing the development of membership inference attacks,this paper explores the main challenges and potential development directions of machine learning privacy protection in the future.

Recently,the fairness of the algorithm has aroused extensive discussion in the machine learning community.Given the widespread popularity of spectral clustering in modern data science,studying the algorithm fairness of spectral clustering is a crucial topic.Existing fair spectral clustering algorithms have two shortcomings:1) poor fairness performance;2) work only for single sensitive attribute.In this paper,the fair spectral clustering problem is regarded as a constrained spectral clustering problem.By solving the feasible solution set of constrained spectral clustering,an unnormalized fair spectral clustering(UFSC) method is proposed to improve fairness performance.In addition,the paper also proposes a fair clustering algorithm suitable for multiple sensitive attribute constraints.Experimental results on multiple real-world datasets demonstrate that the UFSC and MFSC are fairer than the existing fair spectral clustering algorithms.

Artificial intelligence has been integrated into all aspects of people's daily lives with the continuous development of machine learning,especially in the deep learning area.Machine learning models are deployed in various applications,enhancing the intelligence of traditional applications.However,in recent years,research has pointed out that personal data used to train machine learning models is vulnerable to the risk of privacy disclosure.Membership inference attacks(MIAs) are significant attacks against the machine learning model that threatens users' privacy.MIA aims to judge whether user data samples are used to train the target model.When the data is closely related to the individual,such as in medical,financial,and other fields,it directly interferes with the user's private information.This paper first introduces the background knowledge of membership inference attacks.Then,we classify the existing MIAs according to whether the attacker has a shadow model.We also summarize the threats of MIAs in different fields.Also,this paper points out the defense means against MIAs.The existing defense mechanisms are classified and summarized according to the strategies for preventing model overfitting,model-based compression,and disturbance.Finally,this paper analyzes the advantages and disadvantages of the current MIAs and defense mechanisms and proposes possible research directions for future MIAs.

The outbreak of pneumonia caused by novel coronavirus (COVID-19) at the end of 2019 was a major public health emergency in human history. In a short period of time, Chinese medical workers have experienced the gradual understanding, evidence accumulation and clinical practice of the unknown virus. So far, National Health Commission of the People's Republic of China has issued seven trial versions of the "Guidelines for the Diagnosis and Treatment of COVID-19". However, it is difficult for clinicians and laymen to quickly and accurately distinguish the similarities and differences among the different versions and locate the key points of the new version. This paper reports a computer-aided intelligent analysis method based on machine learning, which can automatically analyze the similarities and differences of different treatment plans, present the focus of the new version to doctors, reduce the difficulty in interpreting the "diagnosis and treatment plan" for the professional, and help the general public better understand the professional knowledge of medicine. Experimental results show that this method can achieve the topic prediction and matching of the new version of the program text through unsupervised learning of the previous versions of the program topic with an accuracy of 100%. It enables the computer interpretation of "diagnosis and treatment plan" automatically and intelligently.

Abstract: Bone age assessment has always been one of the key issues and difficulties in forensic science. With the gradual development of machine learning in many industries, it has been widely introduced to imageology, genomics, oncology, pathology, surgery and other medical research fields in recent years. The reason why the above research fields can be closely combined with machine learning, is because the research subjects of the above branches of medicine belong to the computer vision category. Machine learning provides unique advantages for computer vision research and has made breakthroughs in medical image recognition. Based on the advantages of machine learning in image recognition, it was combined with bone age assessment research, in order to construct a recognition model suitable for forensic skeletal images. This paper reviews the research progress in bone age assessment made by scholars at home and abroad using machine learning technology in recent years., Competing Interests: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose., (Copyright© by the Editorial Department of Journal of Forensic Medicine.)

Radiotherapy is one of the main treatments for tumor with increasingly high request for technique precision and the equipment stability. Machine learning may bring radiotherapy simplicity, individualization and precision, and may improve the automatic level of planning and quality assurance. Based on the process of radiotherapy, this paper reviews the applications and researches on machine learning, with an emphasis on deep learning, and proposes the prospects in the following aspects: segmentation of normal tissue and tumor, planning, treatment delivery, quality assurance and prognosis prediction.

Medical claims database is an important source of data for studying the characteristics, and burden of diseases, to provide a basis for the development of policy on management. The database is usually used to identify patients through International Classification of Diseases and free text-building algorithms, thus it is crucial to validate whether the algorithm is correctly identifing the targeted population. This paper introduces both traditional and emerging validation methods including machine learning, natural language processing and database linkage etc .. We also have tried to present a suitable validation method for the current situation in China, so as to promote the application of big data in medical areas and to provide reference for epidemiology studies, based on medical claims database in this country.

Reproducible build is the ability to recreate binary artifacts in a predefined build environment.Due to the role of reproducible build in ensuring the security of software construction environment and improving the efficiency of software construction and distribution,many open source software repositories(such as Debian) have carried out software reproducible build practice.However,due to the lack of sufficient judgment information and the complexity and diversity of source files,it is still a time-consuming and laborious challenge to determine why software can not be built reproducibly.In order to overcome this challenge,this paper studies the classification and detection of software unreproducible build causes based on machine learning.This paper stu-dies four typical reasons for unreproducible build,namely timestamp,fileordering,randomness and locale.This method uses the word vector generated by word2vec to represent the text log,and then cooperates with the logistic regression model to learn and train the text corpus combined with the difference log and the build log,so as to realize the automatic classification of the causes of unreproducible build.In this paper,the algorithm is implemented and tested on 671 unreproducible build Debian software packa-ges.Experimental results show that our method achieves a macro average precision of 80.75% and a macro average recall of 86.07%,which are better than other commonly used machine learning algorithms.In addition,we also analyze the relevance and importance of difference log and build log.Result indicates that both of them are significant for the classification of unreproducible build causes.This method provides a reliable research basis for automatic classification of unreproducible build causes.

With the increasing globalization of football,the global player transfer market is becoming more and more prosperous.However,as the most important factor affecting player transfer transaction,the player’s transfer value lacks in-depth model and application research.In this paper,the FIFA’s official player database is taken as the research object.Firstly,on the premise of distinguishing different player positions,Box-Cox transformation,F-Score feature selection,etc.are used to perform feature processing on the original data set.Secondly,the player value prediction model is constructed by XGBoost,and compared with the main machine learning algorithms such as random forest,AdaBoost,GBDT and SVR for 10-fold cross validation experiments.Experimental results prove that the XGBoost model has a performance advantage over the existing models on the indicators of R2,MAE and RMSE.Finally,on the basis of constructing the value prediction model,this paper integrates the SHAP framework to analyze the important factors affecting the players’ value score in different positions,and provides decision support for some scenarios,such as player’s value score evaluation,comparative analysis,and training strategy formulation,etc.

In the field of natural language processing, named entity recognition is the first key step of information extraction. Named entity recognition task aims to recognize named entities from a large number of unstructured texts and classify them into predefined types. Named entity recognition provides basic support for many natural language processing tasks such as relationship extraction, text summarization, machine translation, etc. This paper first introduces the definition of named entity recognition, research difficulties, particularity of Chinese named entity recognition, and summarizes the common Chinese and English public datasets and evaluation criteria in named entity recognition tasks. Then, according to the development history of named entity recognition, the existing named entity recognition methods are investigated, which are the early named entity recognition methods based on rules and dictionaries, the named entity recognition methods based on statistic and machine learning, and the named entity recognition methods based on deep learning. This paper summarizes the key ideas, advantages and disadvan-tages and representative models of each named entity recognition method, and summarizes the Chinese named entity recognition methods in each stage. In particular, the latest named entity recognition based on Transformer and based on prompt learning are reviewed, which are state-of-the-art in deep learning-based named entity recognition methods. Finally, the challenges and future research trends of named entity recognition are discussed.

One of the most important technologies in modern information and service industry is social influence analysis.More and more researchers in social networks focus on social influence.Real social networks are evolving rather than static.The proposal of evolutionary network also brings new challenges and opportunities.At the same time,the massive social information in the evolutionary network also provides strong support for the rapid development of big data analysis technology.In this paper,evolutionary network and influence maximization are discussed.It also discusses the diffusion model of social influence analysis and the influence analysis method based on social network big data.At the same time,some widely used influence algorithms are further sorted out.In addition,this paper also discusses the relationship between big data,evolutionary networks,and social influence maximization.This paper aims to help other researchers to better understand the existing work and provide new ideas for the influence analysis of social networks through the influence analysis of large-scale social networks.

Federated learning is proposed to solve the contradiction between data sharing and privacy-preserving.It aims to build collaborative models by securely interacting irreversible information (e.g.,model parameters or gradient updates).However,the risks of privacy leakage and malicious attacks in the process of model local training,information interaction and parameter transmission have brought major challenges to the practical application of federated learning.This paper summarizes the Attack beha-viors and corresponding defense strategies in the modeling and deployment process of federated learning.Firstly,this paper briefly reviews the development process of federated learning and the basic modeling process.Next,it classifies attack behaviors in fede-ral learning training and deployment from three aspects:confidentiality,availability and integrity,and combs the latest research on privacy theft and malicious attacks.Then,it summarizes defense countermeasures from two directions: honest-but-curious attac-kers and malicious attackers,and analyzes the defense capabilities of different strategies.Finally,it presents some discussions about the problems and challenges of attack and defense methods in the practice of federated learning.Besides,it looks forward to their future development direction of federated learning in defense strategy and system design.

Acute respiratory distress syndrome (ARDS) is a serious threat to human life and health disease, with acute onset and high mortality. The current diagnosis of the disease depends on blood gas analysis results, while calculating the oxygenation index. However, blood gas analysis is an invasive operation, and can't continuously monitor the development of the disease. In response to the above problems, in this study, we proposed a new algorithm for identifying the severity of ARDS disease. Based on a variety of non-invasive physiological parameters of patients, combined with feature selection techniques, this paper sorts the importance of various physiological parameters. The cross-validation technique was used to evaluate the identification performance. The classification results of four supervised learning algorithms using neural network, logistic regression, AdaBoost and Bagging were compared under different feature subsets. The optimal feature subset and classification algorithm are comprehensively selected by the sensitivity, specificity, accuracy and area under curve (AUC) of different algorithms under different feature subsets. We use four supervised learning algorithms to distinguish the severity of ARDS (P/F ≤ 300). The performance of the algorithm is evaluated according to AUC. When AdaBoost uses 20 features, AUC = 0.832 1, the accuracy is 74.82%, and the optimal AUC is obtained. The performance of the algorithm is evaluated according to the number of features. When using 2 features, Bagging has AUC = 0.819 4 and the accuracy is 73.01%. Compared with traditional methods, this method has the advantage of continuously monitoring the development of patients with ARDS and providing medical staff with auxiliary diagnosis suggestions.

Radar emitter signal deinterleaving is a key technology for radar signal reconnaissance and an essential part of battlefield situational awareness. This paper systematically sorts out the mainstream technology of radar emitter signal deinterleaving. It summarizes the main research progress in radar emitter signal deinterleaving from three directions: interpulse modulation characteristics-based, intrapulse modulation characteristics-based, and machine learning-based research. Particularly, this paper focuses on explaining the principle and technical characteristics of the latest deinterleaving technology, such as neural network-based and data stream clustering-based techniques. Finally, the shortcomings of the current radar emitter deinterleaving technology are summarized, and the future trend is predicted.

Road traffic accidents (RTA) can cause a large number of casualties and property losses. Driving fatigue is one of the important factors leading to RTA. Electrophysiological signals, as a kind of information feedback for the nervous system to regulate body functions, can reflect drivers’ fatigue state. However, there is a lack of systematic reviews on the current research on electrophysiological signals as information input of machine learning methods for driving fatigue recognition. By investigating fatigue-related literature, the current paper summarized the neural regulation mechanism of fatigue, clarified that driving fatigue is caused by both psychological and physiological loads, recognized inducing factors related to driving fatigue, and summed up electrophysiological signals now in use of driving fatigue recognition, as well as their physiological mechanisms and related indicators. Machine learning algorithms are widely used in identifying driving fatigue. Based on existing studies that used electrophysiological signals as information input source and applied various machine learning algorithms to build driving fatigue identification models, this paper compared the effectiveness of various machine learning algorithms, and described the advantages and disadvantages of supervised machine learning. It is pointed out that suitable classification algorithms should be selected according to sample conditions and model eigenvalues when applied to driving fatigue recognition. In addition, a variety of electrophysiological signals as information sources can help improve the accuracy of a fatigue recognition model, but the increase of model input eigenvalues cannot. Finally, the research progress of identification methods based on electrophysiological signals provided new opportunities for identifying driving fatigue.

During daily operation of Substation, if Gas Insulated Switchgear (GIS)equipment breaks down, it will lead to the suspension of substation power supply, and national production and life cannot be guaranteed. This paper discusses how to locate the breakdown of GIS efficiently and effectively. The Distributed Acoustic Sensing (DAS) system described in this paper is combined with Support Vector Machines (SVM) in machine learning method, which can realize fault location and facilitate fault elimination. At the same time, it is of great significance to restore the power supply of substation in time.

Based on blockchain technology,Ethereum Solidity smart contract as a computer protocol is designed to spread,verify,or execute contracts in an informative way,and it provides a foundation for various distributed application services.Although implemented for less than six years,its security problems have frequently broken out and caused substantial financial losses,which attracts more attention in the security inspection research.This paper firstly introduces some specific mechanisms and operating principles of smart contracts based on Ethereum related techniques,and analyzes some smart contract vulnerabilities occurring frequently and deriving from the characteristics of smart contracts.Then,this paper explains the traditional mainstream smart contract vulnerability detecting tools in terms of symbolic execution,fuzzing,formal verification,and taint analysis.In addition,in order to cope with the endless new vulnerabilities and the need to improve the efficiency of detection,vulnerabilities detection based on machine learning in recent years is classified and summarized according to the various ways of problem transformation in three perspectives including text processing,non-Euclidean graph and standard image.Finally,this paper proposes to formulate more extensive and accurate standardized information database and measurement indicators towards the insufficiency of the detection methods in two directions.

Modern compilers offer many optimization options.It is a complex problem to choose which parameter values,which combination of options and in which order to apply these options.Among them,the optimization phase ordering is the most dif-ficult one.With the improvement of traditional methods (iterative compilation combined with heuristic optimization search) and the emergence of new technologies (machine learning),it is possible to build a relatively efficient and intelligent compiler automatic tuning framework.This paper summarizes the research ideas and application methods of predecessors by investigating the related research in the past decades.Firstly,the development of compiler automatic tuning is introduced,including early manual methods,cost function driven methods,iterative compilation and machine learning-based prediction methods.Then,this paper focuses on the direct prediction based on machine learning and the automatic optimization method of iterative compilation driven by machine learning.Last but not least,several successful frameworks and some recent research results are listed.Current challenges and some key future research directions are also pointed out.

The site selection optimization of coal mine safety emergency reserve center is an important foundation for promoting the construction of coal mine safety emergency response system. In order to improve the accuracy and reasonableness of coal mine safety emergency reserve center site selection, it is proposed to establish a machine learning combination model for coal mine safety emergency reserve center site selection, integrating multi-source spatial data by using demographic, transportation, economic and natural factors to improve the accuracy and scientificity of coal mine safety emergency reserve center site selection. The accuracy and scientificity of coal mine safety emergency reserve center site selection are improved. Firstly, the ArcGIS is used to process multi-source spatial data through tasks such as fishing net division, spatial linking and projection respectively, and the SMOTEENN algorithm is utilized to avoid the negative impact of data imbalance, so as to construct the dataset applicable to the analysis of machine learning model. Secondly, by comparing and analyzing different machine learning algorithms, different feature selection methods and different parameter optimization methods, it is concluded that the XGBoost machine learning algorithm, the Boruta algorithm and genetic algorithm have better performance than other machine learning algorithms, feature selection methods and parameter optimization methods in the site selection analysis of coal mine safety and emergency reserve center. Therefore, based on the advantages of each algorithm, this paper obtains a combined machine learning model for coal mine safety emergency reserve center site selection. Finally, the SHAP analysis is introduced to analyze the influence degree and direction of different features to quantitatively assess the contribution of input data in decision-making and enhance the interpretability of the model. The results show that the combined model of coal mine safety emergency reserve center siting has an excellent performance, with 0.976, 0.966, 0.989, 0.977, 0.996 in accuracy, precision, recall, F1 value and AUC, respectively, which can provide a powerful support for siting decision-making, and the model interpretable analysis can also provide a scientific reference for coal mine safety emergency reserve center siting.

【Objective】In this paper, machine learning method is applied to 30 Tbit/s (60 × 500 Gbit/s) Nyquist Dual Polarization-16 Quadrature Amplitude Modulation (DP-16QAM) system after 6 300 km transmission in G. 654E optical fiber. Nonlinear channel equalization is used to reduce the transmission Bit Error Rate (BER).【Methods】Referring to the "receptive field" mechanism of convolution neural network, the size of "convolution core" is designed, and the data set is constructed according to the divided sampling data. The artificial neural network is constructed by optimizing the parameters. The one-to-one data corresponding to the transmission and reception of different wavelengths, different optical signal-to-noise ratios, and different fiber input powers in the C-band are collected. Refer to the classic full-connection neural network structure, the neural network is constructed according to the data structure of the data set. The network fitting is carried out for the real part and the imaginary part respectively. After training stage, the test data is sent into the network, and the performances are compared with the traditional methods.【Results】Two kinds of neural networks are used to fit the transmission BER under 60 different wavelength transmission conditions of C band frequency from 191.562 5 to 195.987 5 THz. Compared with Maximum Likelihood Sequence Estimation (MLSE), Network 1 has an average reduction of 23% in BER, and Network 2 has an average reduction of 41% in BER. A frequency of 193.812 5 THz is then selected for the calculation of the fiber input power ranging from 14 to 19 dBm. The average improvement in network 1 and network 2 are 32% and 52%, respectively. Under different optical signal-to-noise ratios, Network 1 has an average improvement of 30%, and Network 2 has an average improvement of 57%.【Conclusion】The two neural networks have excellent performance in nonlinear equalization of coherent transmission systems. At the same time, the number of network layers and nodes will jointly affect the fitting results. Increasing the number of layers and nodes can obtain better fitting results, but the corresponding parameters, training time and the required space will also increase. Therefore, in the application, the actual situation should be considered to choose between the fitting performances and the model attributes.

Purposes The prediction of drug-target interactions plays a crucial role in drug relocation and drug development. Methods A multi-feature fusion algorithm is proposed based on the combination of Redundancy-Correlation and Interaction (RCI), and the prediction model is built by combining the stacked ensemble classifier. First, high-dimensional features of the drug and target are extracted for multi-feature fusion, and RCI is used to build a non-redundant and relevant interactive feature subset. Then, the feature subset is input into a stacked ensemble classifier composed of multiple base learners for training. Finally, the prediction is carried out in two benchmark drug target networks. Findings The experimental results show that the accuracies of ACC and AUC values of the model in this paper are better than those of the existing benchmark methods, indicating the effectiveness of the proposed algorithm.

To gain a profound understanding of the current status and future trends in the comprehensive prevention and control of coal spontaneous combustion, this paper comprehensively reviews stages and research progress in the aspects of “monitoring and early warning, prevention technology, and emergency rescue”. The following future research aspects are proposed: utilizing advanced technologies such as comprehensive multivariate statistical analysis, the Internet of Things (IoT), machine learning, and artificial intelligence to enhance the accuracy of coal spontaneous combustion prediction and forecasting. This involves constructing an intelligent, proactive warning, and situational judgment mechanism to achieve intelligent prediction, forecasting, and proactive warning systems. Through the comprehensive application of interdisciplinary research methods, gaining in-depth insights into the spontaneous combustion mechanism, innovative fire-resistant materials, and the development of new microbial technologies, theses make the technology of spontaneous combustion prevention and control of coal present a trend of in-depth research and multiple innovation. Under the framework of an efficient, collaborative, and intelligent rescue mechanism, the emergency rescue for coal spontaneous combustion is advancing towards a more intelligent and information-oriented direction. The goal is to promote the intelligent advancement of the coal spontaneous combustion prevention and control level.

Deep learning and neural network models have been new research directions and hot issues in the fields of machine learning and artificial intelligence in recent years. Deep learning has made a breakthrough in the applications of image and speech recognitions, and also has been extensively used in the fields of face recognition and information retrieval because of its special superiority. Bone X-ray images express different variations in black-white-gray gradations, which have image features of black and white contrasts and level differences. Based on these advantages of deep learning in image recognition, we combine it with the research of bone age assessment to provide basic datum for constructing a forensic automatic system of bone age assessment. This paper reviews the basic concept and network architectures of deep learning, and describes its recent research progress on image recognition in different research fields at home and abroad, and explores its advantages and application prospects in bone age assessment., Competing Interests: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose., (Copyright© by the Editorial Department of Journal of Forensic Medicine.)

Generative adversarial networks (GAN) has become a popular research direction in the field of deep learning. The unique adversarial idea of GAN comes from the two-player zero-sum game in game theory. How to solve the problems of unstable GAN training, poor quality of generated samples, inadequate evaluation system, poor interpretability and other issues is critical and difficult in current GAN research. This paper investigates the research background and development trend of GAN. First, the basic idea and algorithm implementation of GAN are described, the advantages and disadvantages of GAN are analyzed, a more systematic classification of existing improved methods is made, and some typical optimization methods and derivative models of GAN are sorted out from two categories based on structure change and loss function variants respectively. Next, the similarities and differences between GAN and other generative models are compared, and their respective advantages and disadvantages are introduced. Then, this paper compares the performance of GAN and its derivative models, and summarizes their operation mechanism, advantages, limitations and applicable scenarios. The applications of GAN in the field of image generation are introduced. Finally, the mainstream evaluation indicators of GAN are listed, the main problems still faced in GAN research are analyzed and corresponding solutions are given. The listed mainstream solutions are compared and analyzed in terms of solution effects and applicability, and the future research direction is prospected.

This paper proposes a user-centered location privacy game mechanism,which is aimed to generate corresponding protection strategies based on the LBS service quality,and reduce the calculation scale and utility loss.This mechanism is based on the Stackelberg game model.When a user requests a LBS service,he/she uses the location ambiguity mechanism to disturb his/her location and send it to the LBS server,making it difficult for the attacker to predict his/her real location.Based on part of their known background knowledge,attackers infer the protection policies of users in the anonymous area and adjust their attack methods to minimize the level of user privacy.In order to solve the problem of large scale and long time calculation by traditional mathematical methods,this paper adopts generating countermeasures network to participate in the generation of protection strategy,and reduces the utility cost as much as possible.The experimental results show that the protection mechanism has good performance in terms of privacy protection level,and at the same time,it significantly reduces the generation time of the protection mechanism while losing some quality of service.

The landslide disaster in China is widespread and serious. Regional landslide risk assessment has always been one of the most important contents of landslide disaster prevention and mitigation. In recent years, with the rapid development of big data and artificial intelligence technology, machine learning technology has gradually been widely used in landslide hazard assessment andachieved good results. Based on a large number of literatures, this paper systematically expounds the research status of landslide risk assessment methods based on machine learning technology. This paper reviews and analyzes the existing research results from three key links: evaluation factor selection and quantization normalization, data cleaning and sample set construction, model selection and training evaluation, and finally puts forward some suggestions on the development trend of machine learning landslide risk evaluation methods.

This paper proposes a four-layer neural network based on randomly feature mapping (FRMFNN) and its fast incremental learning algorithms. First, FRMFNN transforms the original input features into randomly mapped features by certain randomly mapping algorithm and stores them in its nodes of first hidden layer. Then, the FRMFNN generates its nodes of second hidden layer using non-linear activation function on all random mapping features. Finally, the second hidden layer is linked to the output layer through the output weights. Since the weights of the first and the second hidden layers are randomly generated according to certain continuous sampling probability distribution, without the updates of the weights, and the output weights can be quickly solved by the ridge regression, avoiding time-consuming training process of the traditional back propagation neural networks. When FRMFNN can't reach the prescribed accuracy, its performance can be continuously improved by its rapid incremental algorithm, thereby avoiding retraining the whole network. This paper, a detail introduction of proposed FRMFNN and its incremental algorithms is provided. What's more, a proof of universal approximation property of FRMFNN is also given. Compared with broad learning system (BLS) and the incremental learning of extreme learning machine (ELM), the experimental results on several popular classification and regression datasets demonstrate the effectiveness of the proposed FRMFNN and its incremental learning algorithms.

The automatic classification of epileptic electroencephalogram(EEG)is significant in the diagnosis and therapy of epilepsy.A classification algorithm for epileptic EEG based on wavelet multiscale analysis and extreme learning machine(ELM)is proposed in this paper.Firstly,wavelet multiscale analysis is applied to the original EEG to extract its sub-bands.Then,two nonlinear methods,i.e.Hurst exponent(Hurst)and sample entropy(SamEn)are used to the feature extraction of EEG and its sub-bands.Finally,ELM algorithm is employed in epileptic EEG classification with the nonlinear features.The proposed method in this paper achieved 99.5% classification accuracy for the discrimination between epileptic ictal and interictal EEG.The result implies that this method has good prospects in the diagnosis and therapy of epilepsy.

With the increasing number of electrocardiogram(ECG)data,extensive application requirements of computer-aided ECG analysis have occurred.In the paper,we propose a variety of strategies to improve the performance of clinical ECG classification algorithm based on Lead Convolutional Neural Network(LCNN).Firstly,we obtained two classifiers by using different preprocessing methods and training methods in the study.Then,we applied the multiple output prediction method to both of them independently.Finally,the Bayesian approach was employed to fuse them.Tests conducted using more than 150 000 ECG records showed that the proposed method had an accuracy of 85.04% and the area under receiver operating characteristic curve(AUC)was 0.918 5,which significantly outperforms traditional methods based on feature extraction techniques.

心脑血管疾病是全球主要的致死和致残病因之一。尽管医疗技术不断进步，心脑血管疾病的诊治仍存在许多亟待解决的问题，人工智能与机器学习的出现为应对这些问题提供了新途径。美国心脏学会2024年发布的科学声明从影像分析、心电图判读、床旁监护、可穿戴设备、基因组学及电子健康记录等方面介绍了人工智能与机器学习在临床中的应用及面临的挑战。本文对该指南进行介绍，并分析人工智能与机器学习在脑血管病领域中应用的优势，为我国未来在这一领域的研究提供参考。 Abstract: Cardiovascular and cerebrovascular diseases are among the leading causes of death and disability worldwide. Despite continuous advancements in medical technology, the diagnosis and treatment of these diseases still face many unresolved challenges. The emergence of artificial intelligence and machine learning provides new approaches to addressing these issues. The American Heart Association’s 2024 scientific statement highlights the applications and challenges of artificial intelligence and machine learning in clinical scenarios, including imaging analysis, electrocardiogram interpretation, bedside monitoring, wearable devices, genomics, and electronic health records. This paper aims to introduce these aspects and analyze the advantages of artificial intelligence and machine learning in the field of cerebrovascular diseases, providing a reference for the future development of artificial intelligence and machine learning in this area in China.

In recent years, many new modulation and multiplexing technologies and dynamic network concepts have been proposed to adapt to the ever-increasing network bandwidth and quality requirements. Network control platform has a systematic and intelli-gent development trend, which requires network managers to constantly monitor the parameters of the network and optimize the network state. However, it is not feasible to arrange additional monitoring equipment in a large range to obtain parameter informa-tion from the perspective of cost control. It is better to use known data and special algorithms to monitor and optimize network per-formance. Machine learning methods are increasingly adopted by the academic community because they are accurate and efficient enough to accomplish these tasks. This paper first reviews the different application scenarios of machine learning algorithms in op-tical network monitoring and optimization tasks. Then it reviews the research achievements in this field, and puts forward the ex-isting problems of machine learning-based optical network monitoring and optimization methods as well as the possible direction of future research. The optical performance monitoring based on machine learning includes failure identification, quality of transmis-sion estimation and channel power prediction. The network configuration optimization method based on machine learning includes reinforcement learning to optimize channel power. For future research direction, we believe that it is possible for researchers to use real data from network operators, newly collected data to dynamically train the model, and transfer learning and data enhancement techniques to ensure the robustness and generalization ability of the algorithm.

In the traditional well log depth matching tasks, manual adjustments are required, which means significantly labor-intensive for multiple wells, leading to low work efficiency. This paper introduces a multi-agent deep reinforcement learning (MARL) method to automate the depth matching of multi-well logs. This method defines multiple top-down dual sliding windows based on the convolutional neural network (CNN) to extract and capture similar feature sequences on well logs, and it establishes an interaction mechanism between agents and the environment to control the depth matching process. Specifically, the agent selects an action to translate or scale the feature sequence based on the double deep Q-network (DDQN). Through the feedback of the reward signal, it evaluates the effectiveness of each action, aiming to obtain the optimal strategy and improve the accuracy of the matching task. Our experiments show that MARL can automatically perform depth matches for well-logs in multiple wells, and reduce manual intervention. In the application to the oil field, a comparative analysis of dynamic time warping (DTW), deep Q-learning network (DQN), and DDQN methods revealed that the DDQN algorithm, with its dual-network evaluation mechanism, significantly improves performance by identifying and aligning more details in the well log feature sequences, thus achieving higher depth matching accuracy.

With the continuous improvement of living standards, people are concerned about not only whether foods are tasty or not, but also the combination of health elements and good flavor. Food flavor components are not only important factors in sensory quality, but also key indicators of the nutritional level of foods. At present, the traditional methods to evaluate and predict food flavor components are time-consuming and labor-intensive, and unable to handle large amounts of data. In contrast, machine learning (ML), the core of artificial intelligence, has incomparable advantages over traditional analytical techniques in distinguishing differences and finding commonalities, and has found good application in the field of food flavor analysis. In this context, this paper focuses on the current research status of ML in the field of food flavor, and introduces the principles and advantages of commonly used ML methods, as well as their latest applications and prospects in food flavor prediction and regulation. It also focuses on the advantages and future trends of modern intelligent sensory evaluation techniques combined with ML in the field of food flavor analysis, with a view to providing new ideas and theoretical foundations for food flavor analysis and prediction.

Significance Impregnated diamond bits (IDBs) have been widely used in various hard rock drilling activities, especially deep drilling. Drilling process with IDBs relies on diamond edge wrapped by metal matrices to break the rock formation, which means the wear pattern of the diamond bit reflects the interaction between bits and rock, as well as the existence form of broken particles, which directly determine the drilling efficiency and service life of the drilling tools. But up to now there have not been so much in-depth study on its wear mechanism and analysing method in geological drilling. Progress and Analysis Therefore, this paper reviews the research literature on IDBs wear in the field of geological drilling, summarizes the abnormal bit wear problem according to engineering project experineces and industrial standards. Firstly, based on the experiences of geological drilling activities and industry standards, different types of abnormal bit wear of drill bits were summarized. Then, the evaluation methods of drill bit wear were introduced from the aspects of wear images and drilling signals. The wear mechanism and influencing factors of the drill bit in the drilling process were sorted out from three aspects: The overall drill bit, diamond, and metal matrix. In addition, mainstream methods of controlling drill bit wear performance were listed, and the research progress of wear analysis equations was introduced, besides, how machine learning methods can assist in the study of drill bit wear was discussed. Conclusion and Prospect Thus, IDBs have great potentials in deep hard rock drilling, and their wear performance is the key influnencing the whole drilling activity. To this end, research directions such as artificial intelligence assisting data analysis, microscale computational simulation, additive manufacturing, and material processing modification were analyzed and discussed, aiming to explore advanced methods for monitoring, analyzing, and regulating the wear of diamond drill bits, so as to meet the needs of remote monitoring and controlling in drilling activities such as geological deep drilling and extraterrestrial drilling.

临床预测模型可基于从群体获得的效应证据去预测个体的结局或风险，是链接循证医学和精准医学的强有力工具。随着数据测量、存储、互联互通及分析技术的不断进步，临床预测模型的应用前景将越来越明朗，但与此同时，也面临着诸多问题和挑战。本文对临床预测模型的基础概念、应用场景以及现在面临的困境和发展机遇做一个梳理。 Abstract: Clinical prediction models can predict individual outcomes or risks based on effect evidence obtained from the population, serving as a powerful tool linking evidence-based medicine and precision medicine. With the continuous advancement of data measurement, storage, interoperability, and analytical techniques, the application prospect for clinical prediction models is becoming increasingly clear. However, it also faces many problems and challenges. This paper aims to provide an overview of the fundamental concepts, application scenarios, current dilemmas, and development opportunities of clinical prediction models.