环境微生物研究中机器学习算法及应用

微生物在环境中无处不在，它们不仅是生物地球化学循环和环境演化的关键参与者，也在环境监测、生态治理和保护中发挥着重要作用。随着高通量技术的发展，大量微生物数据产生，运用机器学习对环境微生物大数据进行建模和分析，在微生物标志物识别、污染物预测和环境质量预测等领域的科学研究和社会应用方面均具有重要意义。机器学习可分为监督学习和无监督学习2大类。在微生物组学研究当中，无监督学习通过聚类、降维等方法高效地学习输入数据的特征，进而对微生物数据进行整合和归类。监督学习运用有特征和标记的微生物数据集训练模型，在面对只有特征没有标记的数据时可以判断出标记，从而实现对新数据的分类、识别和预测。然而，复杂的机器学习算法通常以牺牲可解释性为代价来重点关注模型预测的准确性。机器学习模型通常可以看作预测特定结果的“黑匣子”，即对模型如何得出预测所知甚少。为了将机器学习更多地运用于微生物组学研究、提高我们提取有价值的微生物信息的能力，深入了解机器学习算法、提高模型的可解释性尤为重要。本文主要介绍在环境微生物领域常用的机器学习算法和基于微生物组数据的机器学习模型的构建步骤，包括特征选择、算法选择、模型构建和评估等，并对各种机器学习模型在环境微生物领域的应用进行综述，深入探究微生物组与周围环境之间的关联，探讨提高模型可解释性的方法，并为未来环境监测、环境健康预测提供科学参考。     

人工智能大模型医学应用研究

近年来,以自然语言处理和视频图像分析为主的人工智能大模型技术得到快速发展,其基本特征是聚焦相关应用领域的共性需求,通过大数据、强算力和复杂算法的高效协同与深度融合,构建通用预训练模型,广泛适配下游任务,有力提高模型的处理性能与研发效率.因此,大模型技术为医学人工智能高质量发展提供了难得契机.本文通过全面梳理国内外大模型的研究进展、关键技术与核心算法,分析总结生物医学领域一系列标准数据集和预训练模型的发展特点,结合医学人工智能的研发实践,深入剖析医学领域大模型构建的应用需求、解决思路与研发经验,助力推动医学大模型创新发展.     

京津冀地区城市化与生态环境交互耦合关系定量测度

城市化与生态环境之间客观上存在着极其复杂的交互耦合关系,如何实现城市化与生态环境协调发展将是世界经济社会发展的核心议题,也是近年来国内外研究的热点命题。首先构建了城市化和生态环境系统综合评价指标体系,然后借助物理学耦合模型,构建了城市化与生态环境动态耦合协调度模型,定量分析了1980—2011年京津冀地区城市化与生态环境的耦合过程与演进趋势。结果表明:人口城市化和生态压力分别对城市化子系统与生态环境子系统的贡献份额最大,明显高于其他因素;在耦合协调度测算模型中,城市化子系统与生态环境子系统3种不同贡献份额所得出的耦合协调度的变化趋势是一致的,表明耦合协调度模型受城市化与生态环境子系统贡献份额比例的影响很小;1980年以来京津冀地区的城市化与生态环境耦合协调度呈现出S型曲线变化,协调类型从严重不协调-城市化受阻发展到高级协调-生态环境滞后类型;正确认识城市化与生态环境交互胁迫的时空动态耦合规律,采取恰当的区域发展政策和适当的城市发展战略,对进一步加快区域城市化进程,改善生态环境,实现京津冀地区城市化与生态环境的协调和可持续发展具有重要的指导意义。     

城市化与生态环境协调发展的动态耦合模型及其在干旱区的应用

在分析城市化进程与生态环境状况之间交互胁迫、动态演进关系的基础上,借助于系统科学理论建立了城市化与生态环境协调发展的动态耦合模型。认为整个城市化过程就是城市化的各个层面与生态环境的综合协调、交互胁迫的耦合发展过程。在这一过程中,城市化与生态环境协调发展系统的演化周期将经历低级协调共生、协调发展、极限发展和螺旋式上升4个阶段,城市化与生态环境协调发展耦合过程的实质,就是系统各要素相互作用、相互制约,由低级协调共生向高级协调发展的螺旋式上升的过程。以河西走廊为例,对干旱区城市化与生态环境协调发展的动态耦合规律进行了实证分析,认为1985~2003年间,河西走廊城市化水平不断提高,生态环境状况曲折下降,生态环境对城市化的响应较为明显,但相对于城市化进程,生态环境的响应又有一定的滞后性。目前河西走廊处于城市化与生态环境的协调发展阶段,然而协调耦合度的增长很快,整个协调发展系统即将进入极限发展阶段,生态环境危机正在进入潜伏期。基于动态耦合模型所建立的协调耦合度,能够较好地反映城市化与生态环境的交互胁迫、动态耦合的演变情况。根据河西走廊耦合度的变化可知,在城市化发展的初期,往往是需要一定的生态破坏为代价的,然而随着城市化的发展,生态环境必将随城市化而好转。因此,正确认识城市化与生态环境交互胁迫的动态耦合规律,采取适当的城市化发展模式,对于促进河西走廊城市化与生态环境的协调发展具有重大意义。     

宁东能源化工基地生态环境脆弱性评价研究

以宁东能源化工基地为研究对象, 生态环境脆弱性评价为研究目标。依据主客观因素, 利用主成分分析法选取 11 个符合宁东基地实情的评价指标, 构建生态环境脆弱性评价的 SVM 模型。将 GA 嵌入 SVM 中以优化其结构及参数, 利用 SVM 良好的泛化能力和高精度的分类性能进行评价。再利用已有的 BP-ANN 模型进行评价并加以比较分析。结果表明 : 宁东基地的生态环境脆弱性为Ⅱ级 , 属于中度脆弱 ; SVM 模型简单、通用、精度高, 可在生态环境脆弱性评价中推广应用。最后对脆弱性形成的原因进行了分析。     

大数据在生态环境领域的应用进展与展望

随着互联网和移动通讯技术的发展,生态环境领域从信息采集到加工处理也进入信息化和数字化时代,数据量呈现爆发式增长,生态环境大数据受到越来越多的关注.生态环境大数据是在对生态环境要素“空天地一体化”连续观测的基础上,集成海量的多源多尺度信息,借助云计算、人工智能及模型模拟等大数据分析技术,实现生态环境大数据的集成分析和信息挖掘.生态环境大数据存在数据来源多样、涉及部门广;数据采集方式不统一;服务对象众多、对专业化服务要求高等特点.大数据已在生态环境领域得到了初步应用,如在全球气候变化预测、生态网络观测与模拟和区域大气污染治理等方面作用明显.目前我国生态环境大数据的发展还存在诸多问题,包括数据共享难、监测技术落后、传感器等监测设备严重依赖进口、数据集成和深度分析能力不足等.随着大数据技术的进步,未来大数据在解决生态环境健康问题、提高重大生态环境风险预警预报水平、提高生态环境领域科学研究水平等方面都将发挥巨大作用.大数据将最终实现生态环境管理决策定量化、精细化,生态环境信息服务多样化、专业化和智能化,为中国社会经济可持续发展和生态文明建设提供技术保障.     

基于熵准则的鲁棒的RBF谷胱甘肽发酵建模

在谷胱甘肽的发酵过程建模中, 当试验数据含有噪音时, 往往会导致模型预测精度和泛化能力的下降。针对该问题, 提出了一种新的基于熵准则的RBF神经网络建模方法。与传统的基于MSE准则函数的建模方法相比, 新方法能从训练样本的整体分布结构来进行模型参数学习, 有效地避免了传统的基于MSE准则的RBF网络的过学习和泛化能力差的缺陷。将该模型应用到实际的谷胱甘肽发酵过程建模中, 实验结果表明: 该方法具有较高的预测精度、泛化能力和良好的鲁棒性, 从而对谷胱甘肽的发酵建模有潜在的应用价值。     

基于熵准则的鲁棒的RBF谷胱甘肽发酵建模

在谷胱甘肽的发酵过程建模中, 当试验数据含有噪音时, 往往会导致模型预测精度和泛化能力的下降。针对该问题, 提出了一种新的基于熵准则的RBF神经网络建模方法。与传统的基于MSE准则函数的建模方法相比, 新方法能从训练样本的整体分布结构来进行模型参数学习, 有效地避免了传统的基于MSE准则的RBF网络的过学习和泛化能力差的缺陷。将该模型应用到实际的谷胱甘肽发酵过程建模中, 实验结果表明: 该方法具有较高的预测精度、泛化能力和良好的鲁棒性, 从而对谷胱甘肽的发酵建模有潜在的应用价值。     

2010年来黄土高原景观生态研究进展

严重的水土流失以及不合理的土地利用加剧了黄土高原土地资源的退化,导致该地区生态环境脆弱、生态系统服务不断下降。针对黄土高原地区存在的问题,我国学者基于景观生态学原理和方法,围绕"景观格局演变-驱动机制-水土流失过程-生态系统服务"的框架开展了大量研究,取得了一系列研究成果。通过梳理和总结2010年以来黄土高原地区景观生态学研究的现状和特点,指出了目前研究中存在的问题和不足,突出表现在区域比较研究、景观格局与生态过程耦合研究、生态服务权衡方法和模型构建等方面比较缺乏。建议未来黄土高原的景观生态学研究应加强区域尺度上的综合研究和不同地区之间的比较研究,深化景观格局演变的形成机理;进一步开展景观格局与过程的定量识别方法学研究,开发格局-过程耦合模型;加强生态系统过程与服务研究,同时开展相应的实证性研究,研发适宜的生态服务权衡模型,进而深入探讨区域生态系统服务的权衡机制。     

BP神经网络在农产品生产与检测中的应用

人工神经网络是人工智能领域中发展迅速的信息处理技术之一,充分发挥人工神经网络的技术优势,是在农业领域内实现生产劳动自动化的重要途径．本文对BP网络模型及其算法进行了分析研究,从农产品的外观评判、生产预测建模和分类分级鉴定等方面综述了国内外最新研究进展,并展望了今后的应用前景。     

Artificial intelligence: Deep learning in oncological radiomics and challenges of interpretability and data harmonization

Over the last decade there has been an extensive evolution in the Artificial Intelligence (AI) field. Modern radiation oncology is based on the exploitation of advanced computational methods aiming to personalization and high diagnostic and therapeutic precision. The quantity of the available imaging data and the increased developments of Machine Learning (ML), particularly Deep Learning (DL), triggered the research on uncovering “hidden” biomarkers and quantitative features from anatomical and functional medical images. Deep Neural Networks (DNN) have achieved outstanding performance and broad implementation in image processing tasks. Lately, DNNs have been considered for radiomics and their potentials for explainable AI (XAI) may help classification and prediction in clinical practice. However, most of them are using limited datasets and lack generalized applicability. In this study we review the basics of radiomics feature extraction, DNNs in image analysis, and major interpretability methods that help enable explainable AI. Furthermore, we discuss the crucial requirement of multicenter recruitment of large datasets, increasing the biomarkers variability, so as to establish the potential clinical value of radiomics and the development of robust explainable AI models.     

湖泊生态环境损害基线判定技术综述

近年来,污染事故引发的湖泊生态环境损害问题屡见不鲜,特别是氮磷营养盐富集导致的湖泊富营养化。因此,对湖泊生态环境损害程度进行科学合理评判,制定湖泊生态环境损害判定规范程序,形成湖泊生态环境损害判定技术势在必行。目前,国内外学者一致认为,确定切实合理的生态环境基线是对生态环境损害进行科学有效评估的关键技术环节和重要前提。基于相关国家标准以及文献调研,对湖泊生态环境基线判定的原则、判定程序以及判定方法进行系统梳理与总结。详细介绍了针对湖泊不同受体的模型推算法在生态环境损害基线判定中的应用,包括相应的判定方法、判定过程以及优缺点。此外,还针对国内外典型湖泊生态环境损害基线判定研究实例进行阐述。最后对湖泊生态环境基线判定工作的未来发展方向进行了展望,以期为形成统一、合理、有效且具有高度普适性的湖泊生态环境基线制定方法与流程提供依据。     

AI applications to medical images: From machine learning to deep learning

PurposeArtificial intelligence (AI) models are playing an increasing role in biomedical research and healthcare services. This review focuses on challenges points to be clarified about how to develop AI applications as clinical decision support systems in the real-world context.MethodsA narrative review has been performed including a critical assessment of articles published between 1989 and 2021 that guided challenging sections.ResultsWe first illustrate the architectural characteristics of machine learning (ML)/radiomics and deep learning (DL) approaches. For ML/radiomics, the phases of feature selection and of training, validation, and testing are described. DL models are presented as multi-layered artificial/convolutional neural networks, allowing us to directly process images. The data curation section includes technical steps such as image labelling, image annotation (with segmentation as a crucial step in radiomics), data harmonization (enabling compensation for differences in imaging protocols that typically generate noise in non-AI imaging studies) and federated learning. Thereafter, we dedicate specific sections to: sample size calculation, considering multiple testing in AI approaches; procedures for data augmentation to work with limited and unbalanced datasets; and the interpretability of AI models (the so-called black box issue). Pros and cons for choosing ML versus DL to implement AI applications to medical imaging are finally presented in a synoptic way.ConclusionsBiomedicine and healthcare systems are one of the most important fields for AI applications and medical imaging is probably the most suitable and promising domain. Clarification of specific challenging points facilitates the development of such systems and their translation to clinical practice.     

Artificial intelligence in drug design

Thanks to the fast improvement of the computing power and the rapid development of the computational chemistry and biology, the computer-aided drug design techniques have been successfully applied in almost every stage of the drug discovery and development pipeline to speed up the process of research and reduce the cost and risk related to preclinical and clinical trials. Owing to the development of machine learning theory and the accumulation of pharmacological data, the artificial intelligence (AI) technology, as a powerful data mining tool, has cut a figure in various fields of the drug design, such as virtual screening, activity scoring, quantitative structure-activity relationship (QSAR) analysis, de novo drug design, and in silico evaluation of absorption, distribution, metabolism, excretion and toxicity (ADME/T) properties. Although it is still challenging to provide a physical explanation of the AI-based models, it indeed has been acting as a great power to help manipulating the drug discovery through the versatile frameworks. Recently, due to the strong generalization ability and powerful feature extraction capability, deep learning methods have been employed in predicting the molecular properties as well as generating the desired molecules, which will further promote the application of AI technologies in the field of drug design.     

Examining the relationship between urbanization and the eco-environment using a coupling analysis: Case study of Shanghai,China

As a key issue in China’s urban development, urbanization creates increasing pressure on the environment. Thus, a better understanding of the relationship between urbanization and the eco-environment is necessary for Chinese policy makers to realize sustainable urbanization development. Making reference to physical coupling models, we developed a coupling coordination degree model in order to examine the relationship between urbanization and the eco-environment in Shanghai, using data from 1980 to 2013. The comprehensive level of Shanghai’s urbanization process during the study period was estimated using an index composed of four primary indicators, namely: demographic urbanization, spatial urbanization, social urbanization, and economic urbanization. We also developed an index system for the eco-environment, which was based on four primary indicators: the environmental level, eco-environmental endowment, eco-environmental pressure and eco-environment response. The entropy method was subsequently employed in order to identify the contribution made by each indicator to the compound system during the study period. The results show that: (1) economic urbanization and eco-environmental response made the greatest contributions to the urbanization subsystem and the eco-environmental subsystem, respectively—these are thus the key factors to consider in policy decisions aiming to adjust the coupling coordination degree between the two subsystems; (2) the two parameters α-urbanization and β-eco-environment were found to have minimal effect on the coupling coordination system; (3) the coupling coordination between urbanization and the eco-environment produced an S-shaped curve, and both subsystems were found to have evolved from seriously unbalanced development at the start of the study period into superiorly balanced development at the close of the study period. The results of this study hold important implications for efforts to achieve the coordinated development of both urbanization and the eco-environment.     

Artificial intelligence and machine learning for medical imaging: A technology review

Artificial intelligence (AI) has recently become a very popular buzzword, as a consequence of disruptive technical advances and impressive experimental results, notably in the field of image analysis and processing. In medicine, specialties where images are central, like radiology, pathology or oncology, have seized the opportunity and considerable efforts in research and development have been deployed to transfer the potential of AI to clinical applications. With AI becoming a more mainstream tool for typical medical imaging analysis tasks, such as diagnosis, segmentation, or classification, the key for a safe and efficient use of clinical AI applications relies, in part, on informed practitioners. The aim of this review is to present the basic technological pillars of AI, together with the state-of-the-art machine learning methods and their application to medical imaging. In addition, we discuss the new trends and future research directions. This will help the reader to understand how AI methods are now becoming an ubiquitous tool in any medical image analysis workflow and pave the way for the clinical implementation of AI-based solutions.     

Combinational Reasoning of Quantitative Fuzzy Topological Relations for Simple Fuzzy Regions

In recent years, formalization and reasoning of topological relations have become a hot topic as a means to generate knowledge about the relations between spatial objects at the conceptual and geometrical levels. These mechanisms have been widely used in spatial data query, spatial data mining, evaluation of equivalence and similarity in a spatial scene, as well as for consistency assessment of the topological relations of multi-resolution spatial databases. The concept of computational fuzzy topological space is applied to simple fuzzy regions to efficiently and more accurately solve fuzzy topological relations. Thus, extending the existing research and improving upon the previous work, this paper presents a new method to describe fuzzy topological relations between simple spatial regions in Geographic Information Sciences (GIS) and Artificial Intelligence (AI). Firstly, we propose a new definition for simple fuzzy line segments and simple fuzzy regions based on the computational fuzzy topology. And then, based on the new definitions, we also propose a new combinational reasoning method to compute the topological relations between simple fuzzy regions, moreover, this study has discovered that there are (1) 23 different topological relations between a simple crisp region and a simple fuzzy region; (2) 152 different topological relations between two simple fuzzy regions. In the end, we have discussed some examples to demonstrate the validity of the new method, through comparisons with existing fuzzy models, we showed that the proposed method can compute more than the existing models, as it is more expressive than the existing fuzzy models.