Automated Prediction of Cardiorespiratory Deterioration in Patients With Single Ventricle

BackgroundPatients with single-ventricle physiology have a significant risk of cardiorespiratory deterioration between their first and second stage palliation surgeries.ObjectivesThe objective of this study is to develop and validate a real-time computer algorithm that can automatically recognize physiological precursors of cardiorespiratory deterioration in children with single-ventricle physiology during their interstage hospitalization.MethodsA retrospective study was conducted from prospectively collected physiological data of subjects with single-ventricle physiology. Deterioration events were defined as a cardiac arrest requiring cardiopulmonary resuscitation or an unplanned intubation. Physiological metrics were derived from the electrocardiogram (heart rate, heart rate variability, ST-segment elevation, and ST-segment variability) and the photoplethysmogram (peripheral oxygen saturation and pleth variability index). A logistic regression model was trained to separate the physiological dynamics of the pre-deterioration phase from all other data generated by study subjects. Data were split 50/50 into model training and validation sets to enable independent model validation.ResultsOur cohort consisted of 238 subjects admitted to the cardiac intensive care unit and stepdown units of Texas Children’s Hospital over a period of 6 years. Approximately 300,000 h of high-resolution physiological waveform and vital sign data were collected using the Sickbay software platform (Medical Informatics Corp., Houston, Texas). A total of 112 cardiorespiratory deterioration events were observed. Seventy-two of the subjects experienced at least 1 deterioration event. The risk index metric generated by our optimized algorithm was found to be both sensitive and specific for detecting impending events 1 to 2 h in advance of overt extremis (receiver-operating characteristic curve area: 0.958; 95% confidence interval: 0.950 to 0.965).ConclusionsOur algorithm can provide 1 to 2 h of advanced warning for 62% of all cardiorespiratory deterioration events in children with single-ventricle physiology during their interstage period, with only 1 alarm being generated at the bedside per patient per day.     

基于数据挖掘探讨徐浩教授治疗冠心病的用药规律

目的分析徐浩教授治疗冠心病的处方用药规律和经验。方法从门诊病历中选取并整理徐浩教授治疗冠心病处方543首,提取中药处方数据,建立标准化医案数据库,利用古今医案云平台系统的用药统计分析、聚类分析、复杂网络分析等数据挖掘功能,分析医案中处方的用药频次、性味归经、常用药对及核心处方等数据。结果整理处方543首,涉及药物217种,治疗冠心病的高频中药有黄芪、丹参、柴胡、炙甘草、三七粉、瓜蒌等,药性多属温、微寒、平,药味多属甘、苦味,主要归经为肝、肺、心经,功效以活血祛瘀、利水消肿、生津养血为主。常用药对有黄芪-丹参、丹参-黄芪、黄芪-三七粉等。聚类分析可将药物分为4类,分别为活血祛瘀止痛,调畅气机,宣痹通阳,益气养血、健脾化痰。复杂网络分析得出核心药物为黄芪、丹参、三七粉、瓜蒌、柴胡、黄芩、炙甘草、天麻、枳壳、制何首乌、党参、延胡索、川芎、薤白。结论徐浩教授治疗冠心病药物多为温、平、甘、苦,归肝、肺、心经,治疗多气血同治、寒热同调、虚实兼顾、注重调神,以温阳补气、活血化瘀、理气化痰、安神定志为法。     

基于数据挖掘分析中医治疗老年肛周会阴部坏死性筋膜炎术后用药规律

目的 基于数据挖掘分析中医治疗老年肛周会阴部坏死性筋膜炎术后的用药规律,指导临床用药.方法 检索中文文献库,纳入相关文献.根据分层聚类、关联规则的数据挖掘方法,对高频用药及组方规律进行分析.结果 纳入文献中,有31篇涉及口服方剂,24篇涉及外洗方剂,总共涉及处方107份.高频口服药19味,药对18对,聚类方4则;高频外...     

An assessment of the risk factors for vitamin D deficiency using a decision tree model

Background and objectivesVitamin D (25-hydroxyvitamin D or 25OHD) has a key role in the pathogenesis of several chronic disorders. Vitamin D deficiency is a common global public health problem. We aimed to evaluate the risk factors associated with vitamin D deficiency using a decision tree algorithm.MethodsA total of 988 adolescent girls, aged 12–18 years old, were recruited to the study. Demographic characteristics, serum biochemical factors, all blood count parameters and trace elements such as Zinc, Copper, Calcium and SOD were measured. Serum levels of vitamin D below 20 ng/ml were considered to be deficiency. 70% of these girls (618 cases) were randomly allocated to a training dataset for the constructing of the decision-tree. The remaining 30% (285 cases) were used as the testing dataset to evaluate the performance of decision-tree. In this model, 14 input variables were included: age, academic attainment of their father, waist circumference, waist to hip ratio, zinc, copper, calcium, SOD, FBG, HDL-C, RBC, MCV, MCHC, HCT. The validation of the model was assessed by constructing a receiver operating characteristic (ROC) curve.ResultsThe results showed that serum Zn concentration was the most important associated risk factor for vitamin D deficiency. The sensitivity, specificity, accuracy and the area under the ROC curve (AUC) values were 79.3%, 64%, 77.8% and 0.72 respectively using the testing dataset.ConclusionsThe results suggest that the serum levels of Zn is an important associated risk factor for identifying subjects with vitamin D deficiency among Iranian adolescent girls.     

From the Cover: Oil sands development contributes polycyclic aromatic compounds to the Athabasca River and its tributaries

For over a decade, the contribution of oil sands mining and processing to the pollution of the Athabasca River has been controversial. We show that the oil sands development is a greater source of contamination than previously realized. In 2008, within 50 km of oil sands upgrading facilities, the loading to the snowpack of airborne particulates was 11,400 T over 4 months and included 391 kg of polycyclic aromatic compounds (PAC), equivalent to 600 T of bitumen, while 168 kg of dissolved PAC was also deposited. Dissolved PAC concentrations in tributaries to the Athabasca increased from 0.009 μg/L upstream of oil sands development to 0.023 μg/L in winter and to 0.202 μg/L in summer downstream. In the Athabasca, dissolved PAC concentrations were mostly <0.025 μg/L in winter and 0.030 μg/L in summer, except near oil sands upgrading facilities and tailings ponds in winter (0.031–0.083 μg/L) and downstream of new development in summer (0.063–0.135 μg/L). In the Athabasca and its tributaries, development within the past 2 years was related to elevated dissolved PAC concentrations that were likely toxic to fish embryos. In melted snow, dissolved PAC concentrations were up to 4.8 μg/L, thus, spring snowmelt and washout during rain events are important unknowns. These results indicate that major changes are needed to the way that environmental impacts of oil sands development are monitored and managed.     

Data Mining Technologies for Blood Glucose and Diabetes Management

Data mining is the process of selecting, exploring, and modeling large amounts of data to discover unknown patterns or relationships useful to the data analyst. This article describes applications of data mining for the analysis of blood glucose and diabetes mellitus data. The diabetes management context is particularly well suited to a data mining approach. The availability of electronic health records and monitoring facilities, including telemedicine programs, is leading to accumulating huge data sets that are accessible to physicians, practitioners, and health care decision makers. Moreover, because diabetes is a lifelong disease, even data available for an individual patient may be massive and difficult to interpret. Finally, the capability of interpreting blood glucose readings is important not only in diabetes monitoring but also when monitoring patients in intensive care units. This article describes and illustrates work that has been carried out in our institutions in two areas in which data mining has a significant potential utility to researchers and clinical practitioners: analysis of (i) blood glucose home monitoring data of diabetes mellitus patients and (ii) blood glucose monitoring data from hospitalized intensive care unit patients.     

Towards Data Warehousing and Mining of Protein Unfolding Simulation Data

Objectives. The prediction of protein structure and the precise understanding of protein folding and unfolding processes remains one of the greatest challenges in structural biology and bioinformatics. Computer simulations based on molecular dynamics (MD) are at the forefront of the effort to gain a deeper understanding of these complex processes. Currently, these MD simulations are usually on the order of tens of nanoseconds, generate a large amount of conformational data and are computationally expensive. More and more groups run such simulations and generate a myriad of data, which raises new challenges in managing and analyzing these data. Because the vast range of proteins researchers want to study and simulate, the computational effort needed to generate data, the large data volumes involved, and the different types of analyses scientists need to perform, it is desirable to provide a public repository allowing researchers to pool and share protein unfolding data.Methods. To adequately organize, manage, and analyze the data generated by unfolding simulation studies, we designed a data warehouse system that is embedded in a grid environment to facilitate the seamless sharing of available computer resources and thus enable many groups to share complex molecular dynamics simulations on a more regular basis.Results.To gain insight into the conformational fluctuations and stability of the monomeric forms of the amyloidogenic protein transthyretin (TTR), molecular dynamics unfolding simulations of the monomer of human TTR have been conducted. Trajectory data and meta-data of the wild-type (WT) protein and the highly amyloidogenic variant L55P-TTR represent the test case for the data warehouse.Conclusions.Web and grid services, especially pre-defined data mining services that can run on or ‘near’ the data repository of the data warehouse, are likely to play a pivotal role in the analysis of molecular dynamics unfolding data. Based on “Grid Warehousing of Molecular Dynamics Protein Unfolding Data”, by Frederic Stahl, Daniel Berrar, Candida Silva, J. Rui Rodrigues, Rui M.M. Brito, and Werner Dubitzky, which appeared in Proceedings of the IEEE/ACM International Symposium on Cluster Computing and the Grid, Cardiff, UK, May 9–12, 2005.     

Structural investigation of ribosomally synthesized natural products by hypothetical structure enumeration and evaluation using tandem MS

Ribosomally synthesized and posttranslationally modified peptides (RiPPs) are a growing class of natural products that are found in all domains of life. These compounds possess vast structural diversity and have a wide range of biological activities, promising a fertile ground for exploring novel natural products. One challenging aspect of RiPP research is the difficulty of structure determination due to their architectural complexity. We here describe a method for automated structural characterization of RiPPs by tandem mass spectrometry. This method is based on the combined analysis of multiple mass spectra and evaluation of a collection of hypothetical structures predicted based on the biosynthetic gene cluster and molecular weight. We show that this method is effective in structural characterization of complex RiPPs, including lanthipeptides, glycopeptides, and azole-containing peptides. Using this method, we have determined the structure of a previously structurally uncharacterized lanthipeptide, prochlorosin 1.2, and investigated the order of the posttranslational modifications in three biosynthetic systems.Ribosomally synthesized and posttranslationally modified peptides (RiPPs) are a major class of natural products as revealed by the genome-sequencing efforts of the past decade (1). RiPPs are biosynthesized from genetically encoded and ribosomally produced precursor peptides, which typically consist of a core peptide that is transformed to the final product and an N-terminal extension called the leader peptide that is usually important for recognition by the posttranslational modification (PTM) enzymes (1). Because of the highly diverse PTMs, these compounds possess vast structural diversity and have a wide range of biological activities, thus representing a fertile ground for exploration. Furthermore, the ribosomal origin of RiPPs makes them particularly well suited for genome mining efforts. By using genome mining to explicitly avoid species harboring biosynthetic gene clusters identical to those that produce known compounds, a combination of strain prioritization and mass spectrometry (MS)-based analysis offers a new route to discovering natural products that can overcome the burden of rediscovery that has increasingly hampered discovery efforts (2, 3). One challenging aspect of high-throughput genome mining for new natural products is the difficulty to determine their molecular structures in high throughput. We present here a method that allows automated RiPP structure elucidation.In contrast to nonribosomal peptides that have an average molecular weight of less than 1,000 Da, as documented in the NORINE database (4), RiPPs in many cases have molecular weights larger than 2,500 Da. Molecules of this size are difficult to rapidly analyze by NMR spectroscopy, rendering MS the most convenient tool for RiPP structural characterization. Even when the precursor peptide sequences are known and the types of PTMs can be predicted based on the sequences of the biosynthetic enzymes (5–8), multiple possible PTM sites on the precursor peptide typically result in a myriad of structures that are often difficult to differentiate. This challenge is further exacerbated by the frequent occurrence of one or more cross-links in RiPPs, which complicates traditional tandem MS-based structure elucidation. One of the main difficulties is that the spectra only contain a small fraction of informative signals among a large number of less diagnostic signals that cloud spectrum interpretation. As the spectra often also vary significantly with different instrument settings (9), selection of the most suitable spectra for drawing conclusions is time consuming and sometimes introduces bias. Indeed, a number of incorrect structural assignments of RiPPs have been reported based on insufficient information content of tandem MS data (10–15). Here, we report use of hypothetical structure enumeration and evaluation (HSEE) for automated and unbiased interpretation of tandem MS data. The method is based on the prediction of a collection of hypothetical structures for a RiPP of certain mass and known biosynthetic information. By listing all of the theoretical daughter ions from this enumeration and automated evaluation of their matches with one or several experimental spectra, the most probable RiPP structure can be determined. We demonstrate here for multiple classes of known RiPPs with complex structures that HSEE is highly effective in analyzing tandem MS data and predicting the correct structure. In addition, we used HSEE to characterize a lanthipeptide whose structure was elusive despite our previous efforts, and to determine the directionality of thiazole-forming enzymes and lanthipeptide synthetases.     

Randomized approximate nearest neighbors algorithm

We present a randomized algorithm for the approximate nearest neighbor problem in d-dimensional Euclidean space. Given N points {x(j)} in R(d), the algorithm attempts to find k nearest neighbors for each of x(j), where k is a user-specified integer parameter. The algorithm is iterative, and its running time requirements are proportional to T·N·(d·(log d) + k·(d + log k)·(log N)) + N·k(2)·(d + log k), with T the number of iterations performed. The memory requirements of the procedure are of the order N·(d + k). A by-product of the scheme is a data structure, permitting a rapid search for the k nearest neighbors among {x(j)} for an arbitrary point x ∈ R(d). The cost of each such query is proportional to T·(d·(log d) + log(N/k)·k·(d + log k)), and the memory requirements for the requisite data structure are of the order N·(d + k) + T·(d + N). The algorithm utilizes random rotations and a basic divide-and-conquer scheme, followed by a local graph search. We analyze the scheme's behavior for certain types of distributions of {x(j)} and illustrate its performance via several numerical examples.