Inflammablog: peer-to-peer online learning in immunology

Is it possible for students in different courses, at different academic levels, and at different universities to learn immunology together using the Internet? We teach a colloquium on inflammation for undergraduates at the University of Arizona and a lecture course on human immunology for graduate students and clinical and basic science fellows at the University of Colorado Anschutz Medical Campus. Students in these programs, being scattered about large campuses, have little time for student-directed discussion and peer interactions, and they never have the opportunity to meet students in the course in the other state. Instead of requiring the usual essays and term papers, we set up a blog (an online discussion group) for the two courses, and required all students to post, and comment on other posts, within and between the courses. Student writing is normally directed at a single reader, the instructor, which seems like a waste of talent; we encouraged peer exchanges. Furthermore, we were interested in observing the interactions between the Colorado students, who were older and sometimes experienced professionals, and the younger Arizonans. We used a blog because it is administratively impossible to enroll the students in two universities in a single courseware (learning management system) site. Blogging has offered insights into students’ comfort with this form of social medium, and into the potential for this approach in light of the rapid adoption of blended and massively open online courses.     

Relational machine learning for electronic health record-driven phenotyping

ObjectiveElectronic health records (EHR) offer medical and pharmacogenomics research unprecedented opportunities to identify and classify patients at risk. EHRs are collections of highly inter-dependent records that include biological, anatomical, physiological, and behavioral observations. They comprise a patient’s clinical phenome, where each patient has thousands of date-stamped records distributed across many relational tables. Development of EHR computer-based phenotyping algorithms require time and medical insight from clinical experts, who most often can only review a small patient subset representative of the total EHR records, to identify phenotype features. In this research we evaluate whether relational machine learning (ML) using inductive logic programming (ILP) can contribute to addressing these issues as a viable approach for EHR-based phenotyping.MethodsTwo relational learning ILP approaches and three well-known WEKA (Waikato Environment for Knowledge Analysis) implementations of non-relational approaches (PART, J48, and JRIP) were used to develop models for nine phenotypes. International Classification of Diseases, Ninth Revision (ICD-9) coded EHR data were used to select training cohorts for the development of each phenotypic model. Accuracy, precision, recall, F-Measure, and Area Under the Receiver Operating Characteristic (AUROC) curve statistics were measured for each phenotypic model based on independent manually verified test cohorts. A two-sided binomial distribution test (sign test) compared the five ML approaches across phenotypes for statistical significance.ResultsWe developed an approach to automatically label training examples using ICD-9 diagnosis codes for the ML approaches being evaluated. Nine phenotypic models for each ML approach were evaluated, resulting in better overall model performance in AUROC using ILP when compared to PART (p = 0.039), J48 (p = 0.003) and JRIP (p = 0.003).DiscussionILP has the potential to improve phenotyping by independently delivering clinically expert interpretable rules for phenotype definitions, or intuitive phenotypes to assist experts.ConclusionRelational learning using ILP offers a viable approach to EHR-driven phenotyping.     

Metabolite selection for machine learning in childhood brain tumour classification

MRS can provide high accuracy in the diagnosis of childhood brain tumours when combined with machine learning. A feature selection method such as principal component analysis is commonly used to reduce the dimensionality of metabolite profiles prior to classification. However, an alternative approach of identifying the optimal set of metabolites has not been fully evaluated, possibly due to the challenges of defining this for a multi-class problem. This study aims to investigate metabolite selection from in vivo MRS for childhood brain tumour classification. Multi-site 1.5 T and 3 T cohorts of patients with a brain tumour and histological diagnosis of ependymoma, medulloblastoma and pilocytic astrocytoma were retrospectively evaluated. Dimensionality reduction was undertaken by selecting metabolite concentrations through multi-class receiver operating characteristics and compared with principal component analysis. Classification accuracy was determined through leave-one-out and k-fold cross-validation. Metabolites identified as crucial in tumour classification include myo-inositol (P < 0.05, $\text{◂,▸}\text{AUC}=0.81\pm 0.01$), total lipids and macromolecules at 0.9 ppm (P < 0.05, $\text{◂,▸}\text{AUC}=0.78\pm 0.01$) and total creatine (P < 0.05, $\text{◂,▸}\text{AUC}=0.77\pm 0.01$) for the 1.5 T cohort, and glycine (P < 0.05, $\text{◂,▸}\text{AUC}=0.79\pm 0.01$), total N-acetylaspartate (P < 0.05, $\text{◂,▸}\text{AUC}=0.79\pm 0.01$) and total choline (P < 0.05, $\text{◂,▸}\text{AUC}=0.75\pm 0.01$) for the 3 T cohort. Compared with the principal components, the selected metabolites were able to provide significantly improved discrimination between the tumours through most classifiers (P < 0.05). The highest balanced classification accuracy determined through leave-one-out cross-validation was 85% for 1.5 T ¹H-MRS through support vector machine and 75% for 3 T ¹H-MRS through linear discriminant analysis after oversampling the minority. The study suggests that a group of crucial metabolites helps to achieve better discrimination between childhood brain tumours.     

A novel ensemble machine learning for robust microarray data classification

Microarray data analysis and classification has demonstrated convincingly that it provides an effective methodology for the effective diagnosis of diseases and cancers. Although much research has been performed on applying machine learning techniques for microarray data classification during the past years, it has been shown that conventional machine learning techniques have intrinsic drawbacks in achieving accurate and robust classifications. This paper presents a novel ensemble machine learning approach for the development of robust microarray data classification. Different from the conventional ensemble learning techniques, the approach presented begins with generating a pool of candidate base classifiers based on the gene sub-sampling and then the selection of a sub-set of appropriate base classifiers to construct the classification committee based on classifier clustering. Experimental results have demonstrated that the classifiers constructed by the proposed method outperforms not only the classifiers generated by the conventional machine learning but also the classifiers generated by two widely used conventional ensemble learning methods (bagging and boosting).     

基于机器学习的血细胞分类研究进展

基于机器学习的血细胞分类方法已经引起了人们的广泛重视。本文探讨了近几年基于机器学习的血液细胞分类领域的相关研究成果与进展,对目前各种研究所用到的数据采集、图像预处理、图像分割、特征提取及分类器分类方法所用新技术做出详细的说明与分析。深度学习在机器学习基础上发展而成,因其端到端、高准确度等优势展现出强大发展前景。目前研究趋向于采取深度学习与人工特征提取结合、改进网络结构等新方法不断提高网络模型分类准确度及泛化性。然而,基于机器学习的血细胞分类技术投入临床使用仍存在一些问题与挑战。     

A primer on gene expression and microarrays for machine learning researchers

Data originating from biomedical experiments has provided machine learning researchers with an important source of motivation for developing and evaluating new algorithms. A new wave of algorithmic development has been initiated with the publication of gene expression data derived from microarrays. Microarray data analysis is particularly challenging given the large number of measurements (typically in the order of thousands) that are reported for relatively few samples (typically in the order of dozens). Many data sets are now available on the web. It is important that machine learning researchers understand how data are obtained and which assumptions are necessary in the analysis. Microarray data have the potential to cause significant impact in machine learning research, not just as a rich and realistic source of cases for testing new algorithms, as has been the UCI machine learning repository in the past decades, but also as a main motivation for their development. In this article, we briefly review the biology underlying microarrays, the process of obtaining gene expression measurements, and the rationale behind the common types of analyses involved in a microarray experiment. We outline the main challenges and reiterate critical considerations regarding the construction of supervised learning models that use this type of data. The goal of this article is to familiarize machine learning researchers with data originated from gene expression microarrays.     

机器学习在肺癌VMAT计划中对危及器官剂量预测的可行性

目的:探讨机器学习在肺癌容积旋转调强（VMAT）治疗计划对心脏和肺的剂量体积直方图（DVH）预测的可行性。方法:选取51例肺癌VMAT计划,随机选取其中43例为训练组,剩余8例为验证组。分析训练组中患者的解剖信息与两侧肺V5、V20和心脏V30、V40的相关性。采用机器学习方法,以解剖信息为输入、危及器官（OAR）的DVH为输出,分别构建并训练关于两侧肺以及心脏的人工神经网络模型。将验证组中8例VMAT计划中的解剖信息分别输入到已经构建好的人工神经网络模型,分别预测OAR的DVH。结果:两侧肺V5、V20和心脏V30、V40受自身体积大小影响可忽略,受OAR与靶区的空间相对位置关系影响较大。患侧肺、对侧肺、心脏的人工神经网络结构模型中隐藏层分别含有41、38、34个神经结点,线性回归系数分别为0.994、0.975、0.986。对验证组中患侧肺和对侧肺的V5、V20的预测误差分别为2.70%[±]1.83%、2.84[%±]1.97%和13.7%[±]7.8%、0.72[%±]0.75%,对心脏V30、V40的预测误差分别为3.20[%±]0.63%、2.1[%±]1.5%,仅对侧肺V5的预测值和实际值差异有统计学意义（P<0.05）。结论:采用人工神经网络方法可以对肺癌VMAT计划中解剖信息与OAR的DVH数据进行学习,构建的人工神经网络模型可预测出患侧肺、心脏V25[~]V60和对侧肺V20的DVH数据,可为临床计划设计提供参考。     

A classification-based machine learning approach for the analysis of genome-wide expression data

Three important areas of data analysis for global gene expression analysis are class discovery, class prediction, and finding dysregulated genes (biomarkers). The clinical application of microarray data will require marker genes whose expression patterns are sufficiently well understood to allow accurate predictions on disease subclass membership. Commonly used methods of analysis include hierarchical clustering algorithms, t-, F-, and Z-tests, and machine learning approaches. We describe an approach called the maximum difference subset (MDSS) algorithm that combines classification algorithms, classical statistics, and elements of machine learning and provides a coherent framework. By integrating prediction accuracy, the MDSS algorithm learns the critical threshold of statistical significance (the alpha or P-value), eliminating the arbitrariness of setting a threshold of statistical significance and minimizing the effect of the normality assumptions. To reduce the false positive rate and to increase external validity of the predictive gene set, a jackknife step is used. This step identifies and removes genes in the initial MDSS with low combined predictive utility. The overall MDSS provides a prediction that is less dependent on an arbitrary study design (sample inclusion or exclusion) and should thus have high external validity. We demonstrate that this approach, unlike other published methods, identifies biomarkers capable of predicting the outcome of anthracycline-cytarabine chemotherapy in cases of acute myeloid leukemia. By incorporating two criteria-statistical significance and predictive utility-the approach learns the significance level relevant for a given data set. The MDSS approach can be used with any test and classifier operator pair.     

机器学习在电子病历中应用的可视化分析

为探讨机器学习方法在电子病历领域应用的研究现状、研究热点与前沿,以2000～2022年中国知网数据库和Web of Science核心合集数据库中关于机器学习在电子病历中应用的相关文献为数据来源,运用CiteSpace软件绘制国家/地区、作者、机构、关键词共现以及关键词突现5个方面科学知识图谱进行可视化对比分析,以便了解国内外研究的差异,为该领域的研究和发展提供参考。     

人工智能及机器学习在心血管疾病中的应用

人工智能(AI)及机器学习(ML)因其独特的优势逐渐在医学领域得到了较为广泛的应用。在心血管疾病中,该技术在处理电子病历记录中繁杂的数据,预测分析疾病发展及预后,自动分析和识别心血管影像学及心律失常,发现疾病新亚型等方面已经取得了一定进展。AI及ML在心血管疾病研究中潜力巨大,将会为心血管领域带来全新的突破。     

Using machine learning algorithms to guide rehabilitation planning for home care clients

Background

General Practitioners and community nurses rely on easily accessible, evidence-based online information to guide practice. To date, the methods that underpin the scoping of user-identified online information needs in palliative care have remained under-explored. This paper describes the benefits and challenges of a collaborative approach involving users and experts that informed the first stage of the development of a palliative care website [1].

Method

The action research-inspired methodology included a panel assessment of an existing palliative care website based in Victoria, Australia; a pre-development survey (n = 197) scoping potential audiences and palliative care information needs; working parties conducting a needs analysis about necessary information content for a redeveloped website targeting health professionals and caregivers/patients; an iterative evaluation process involving users and experts; as well as a final evaluation survey (n = 166).

Results

Involving users in the identification of content and links for a palliative care website is time-consuming and requires initial resources, strong networking skills and commitment. However, user participation provided crucial information that led to the widened the scope of the website audience and guided the development and testing of the website. The needs analysis underpinning the project suggests that palliative care peak bodies need to address three distinct audiences (clinicians, allied health professionals as well as patients and their caregivers).

Conclusion

Web developers should pay close attention to the content, language, and accessibility needs of these groups. Given the substantial cost associated with the maintenance of authoritative health information sites, the paper proposes a more collaborative development in which users can be engaged in the definition of content to ensure relevance and responsiveness, and to eliminate unnecessary detail. Access to volunteer networks forms an integral part of such an approach.     

Design of a smart biomarker for bioremediation: a machine learning approach

Many trace elements (TE) occur naturally in marine environments and accomplish decisive functions in humans to maintain good health. Mytilus galloprovincialis (MG) is a rich source of TE, but since it is grown near industrial outfalls, they become polluted with elevated levels of TE concentration and serve as biomarkers of pollution. As bioremediation is increasingly reliant on machine learning data processing techniques, we propose the information theoretic concept of using MG for bioremediation. The in situ bioremediation in MG is accomplished by reduction in concentration of TE by the technique of determinant inequalities and the maximization of Mutual Information (MI) without adding any chemical element externally. We bring out the superiority of our technique of MI over that of Principal Component Analysis (PCA) in predicting lower concentration for bioremediation of Cd and Pb in MG.     

TRM: a powerful two-stage machine learning approach for identifying SNP-SNP interactions

Studies have shown that interactions of single nucleotide polymorphisms (SNPs) may play an important role in understanding the causes of complex disease. We have proposed an integrated machine learning method that combines two machine-learning methods-Random Forests (RF) and Multivariate Adaptive Regression Splines (MARS)-to identify a subset of important SNPs and detect interaction patterns more effectively and efficiently. In this two-stage RF-MARS (TRM) approach, RF is first applied to detect a predictive subset of SNPs, and then MARS is used to identify the interaction patterns. We evaluated the TRM performances in four models. RF variable selection was based on out-of-bag classification error rate (OOB) and variable important spectrum (IS). Our results support that RF(OOB) had better performance than MARS and RF(IS) in detecting important variables. This study demonstrates that TRM(OOB) , which is RF(OOB) plus MARS, has combined the strengths of RF and MARS in identifying SNP-SNP interactions in a scenario of 100 candidate SNPs. TRM(OOB) had greater true positive rate and lower false positive rate compared with MARS, particularly for searching interactions with a strong association with the outcome. Therefore, the use of TRM(OOB) is favored for exploring SNP-SNP interactions in a large-scale genetic variation study.     

A comparison of machine learning techniques for detection of drug target articles

Important progress in treating diseases has been possible thanks to the identification of drug targets. Drug targets are the molecular structures whose abnormal activity, associated to a disease, can be modified by drugs, improving the health of patients. Pharmaceutical industry needs to give priority to their identification and validation in order to reduce the long and costly drug development times. In the last two decades, our knowledge about drugs, their mechanisms of action and drug targets has rapidly increased. Nevertheless, most of this knowledge is hidden in millions of medical articles and textbooks. Extracting knowledge from this large amount of unstructured information is a laborious job, even for human experts. Drug target articles identification, a crucial first step toward the automatic extraction of information from texts, constitutes the aim of this paper. A comparison of several machine learning techniques has been performed in order to obtain a satisfactory classifier for detecting drug target articles using semantic information from biomedical resources such as the Unified Medical Language System. The best result has been achieved by a Fuzzy Lattice Reasoning classifier, which reaches 98% of ROC area measure.