Big data analysis using modern statistical and machine learning methods in medicine

The use of prognostic models in allogeneic transplants: a perspective guide for clinicians and investigators

Allogeneic hematopoietic cell transplant (HCT) can cure many hematologic diseases, but it carries the potential risk of increased morbidity and mortality rates. Prognostic evaluation is a scientific entity at the core of care for potential recipients of HCT. It can improve the decision-making process of transplant vs no transplant, help choose the best transplant strategy and allows for future trials targeting patients' intolerances to transplant; hence, it ultimately improves transplant outcomes. Prognostic models are key for appropriate actuarial outcome estimates, which have frequently been shown to be better than physicians' subjective estimates. To make the most accurate prognostic evaluation for HCT, one should rely on >1 prognostic model. For relapse and relapse-related mortality risks, the refined disease risk index is currently the most informative model. It can be supplemented with disease-specific models that consider genetic mutations as predictors in addition to information on measurable residual disease. For nonrelapse mortality and HCT-related morbidity risks, the HCT-comorbidity index and Karnofsky performance status have proven to be the most reliable and most accepted by physicians. These can be supplemented with gait speed as a measure of frailty. Some other global prognostic models might add additional prognostic information. Physicians' educated perceptions can then put this information into context, taking into consideration conditioning regimen and donor choices. The future of transplant mandates (1) clinical investigators specifically trained in prognostication, (2) increased reliance on geriatric assessment, (3) the use of novel biomarkers such as genetic variants, and (4) the successful application of novel statistical methods such as machine learning.

Predictive modeling the probability of suffering from metabolic syndrome using machine learning: A population-based study

Background

There is an increasing trend of Metabolic syndrome (MetS) prevalence, which has been considered as an important contributor for cardiovascular disease (CVD), cancers and diabetes. However, there is often a long asymptomatic phase of MetS, resulting in not diagnosed and intervened so timely as needed. It would be very helpful to explore tools to predict the probability of suffering from MetS in daily life or routinely clinical practice.

Objective

To develop models that predict individuals' probability of suffering from MetS timely with high efficacy in general population.

Methods

The present study enrolled 8964 individuals aged 40-75 years without severe diseases, which was a part of the REACTION study from October 2011 to February 2012. We developed three prediction models for different scenarios in hospital (Model 1, 2) or at home (Model 3) based on LightGBM (LGBM) technique and corresponding logistic regression (LR) models were also constructed for comparison. Model 1 included variables of laboratory tests, lifestyles and anthropometric measurements while model 2 was built with components of MetS excluded based on model 1, and model 3 was constructed with blood biochemical indexes removed based on model 2. Additionally, we also investigated the strength of association between the predictive factors and MetS, as well as that between the predictors and each component of MetS.

Results

In this study, 2714 (30.3%) participants suffer from MetS accordingly. The performances of the LGBM models in predicting the probability of suffering from MetS produced good results and were presented as follows: model 1 had an area under the curve (AUC) value of 0.993 while model 2 indicated an AUC value of 0.885. Model 3 had an AUC value of 0.859, which is close to that of model 2. The AUC values of LR model 1 and 2 for the scenario in hospital and model 3 at home were 0.938, 0.839 and 0.820 respectively, which seemed lower than that of their corresponding machine learning models, respectively. In both LGBM and logistic models, gender, height and resting pulse rate (RPR) were predictors for MetS. Women had higher risk of MetS than men (OR 8.84, CI: 6.70-11.66), and each 1-cm increase in height indicated 3.8% higher risk of suffering from MetS in people over 58 years, whereas each 1- Beat Per Minute (bpm) increase in RPR showed 1.0% higher risk in individuals younger than 62 years.

Conclusion

The present study showed that the prediction models developed by machine learning demonstrated effective in evaluating the probability of suffering from MetS, and presented prominent predicting efficacies and accuracies. Additionally, we found that women showed a higher risk of MetS than men, and height in individuals over 58 years was important factor in predicting the probability of suffering from MetS while RPR was of vital importance in people aged 40-62 years.

A comprehensive study of data intelligence in the context of big data analytics

Modern systems like the Internet of Things, cloud computing, and sensor networks generate a huge data archive. The knowledge extraction from these huge archived data requires modified approaches in algorithm design techniques. The field of study in which analysis of such huge data is carried out is called big data analytics, which helps to optimize the performance with reduced cost and retrieves the information efficiently. The enhancement of traditional data analytics needs to modify to suit big data analytics because it may not manage huge amounts of data. The real thought is how to design the data mining algorithms suitable to handle big data analysis. This paper discusses data analytics at the initial level, to begin with, the insights about the analysis process for big data. Big data analytics have a current research edge in the knowledge extraction field. This paper highlights the challenges and problems associated with big data analysis and provide inner insights into several techniques and methods used.

Development of a graphical model of causal gene regulatory networks using medical big data and Bayesian machine learning

Background: Data collection from medicine and biomedical science is becoming a large task and increasingly complicated with each passing day. Machine learning methods have been applied to elucidate interactions between genes and genes and their environment.Current Concepts: Many machine learning methods have been used to determine the statistical meaning or relationship in the prediction or progression of diseases through the creation of causal networks based on medical big data. Through these analyses, the occurrence and progression of diseases have been shown to be related to several genes and environmental factors. However, these methods cannot identify the key upstream regulators inferred from genomic, clinical, and environmental medical data.Discussion and Conclusion: The causal Bayesian network (CBN) is a machine learning method that can be used to understand a causal network inferred from the gene expression data. The CBN can help identify the key upstream regulators through examining the causal network inferred from medical big data having genomic information. We can easily improve the clinical outcome through regulation of these identified key upstream factors. Therefore, the CBN may be a powerful and flexible tool in the era of precision medicine.

Multiplicative Holts Winter Model for Trend Analysis and Forecasting of COVID-19 Spread in India

The surge of the novel COVID-19 caused a tremendous effect on the health and life of the people resulting in more than 4.4 million confirmed cases in 213 countries of the world as of May 14, 2020. In India, the number of cases is constantly increasing since the first case reported on January 30, 2020, resulting in a total of 81,997 cases including 2649 deaths as of May 14, 2020. To assist the government and healthcare sector in preventing the transmission of disease, it is necessary to predict the future confirmed cases. To predict the dynamics of COVID-19 cases, in this paper, we project the forecast of COVID-19 for five most affected states of India such as Maharashtra, Tamil Nadu, Delhi, Gujarat, and Andhra Pradesh using the real-time data. Using Holt–Winters method, a forecast of the number of confirmed cases in these states has been generated. Further, the performance of the method has been determined using RMSE, MSE, MAPE, MAE and compared with other standard algorithms. The analysis shows that the proposed Holt–Winters model generates RMSE value of 76.0, 338.4, 141.5, 425.9, 1991.5 for Andhra Pradesh, Maharashtra, Gujarat, Delhi and Tamil Nadu, which results in more accurate predictions over Holt’s Linear, Auto-regression (AR), Moving Average (MA) and Autoregressive Integrated Moving Average (ARIMA) model. These estimations may further assist the government in employing strong policies and strategies for enhancing healthcare support all over India.

GLIMPSE: a glioblastoma prognostication model using ensemble learning-a surveillance, epidemiology, and end results study

Purpose

Glioblastoma is one of the most common and aggressive brain tumors in the world with a poor prognosis. A glioblastoma prognostication model has the potential to improve the cancer's standard of care. No other paper has looked at using ensemble learning with a population database to predict multiple binary glioblastoma survival outcomes.

Methods

We utilized ensemble learning to design, build, and test a prognostication system for glioblastoma for short-, intermediate- and long-term survival, based on various clinical features. We used the population database SEER which covers 17 different registries. The most important prognostic features were identified and used as a clinical feature set. The statistical feature set was determined using Random Forests. The accuracy, sensitivity, specificity, area under the receiver operating characteristic (AUROC), positive predictive value (PPV), and negative predictive value (NPV) were reported.

Results

Statistically-determined feature sets had the best performance. All the top models for short, intermediate, and long-term survival were random forests. With regards to short-term survival, top model had metrics AUROC = 0.937, accuracy = 86%, specificity = 88%, sensitivity = 85%, NPV = 85%, and PPV = 87%. For long-term survival, the top model had AUROC = 0.893, accuracy = 81%, specificity = 79%, sensitivity = 83%, NPV = 82%, and PPV = 79%. The top intermediate-term survival prediction had AUROC ≥ 0.780 and the other metrics were at least 70%.

Conclusions

Our ensemble models were high-performing and achieved AUROCs as high as 0.94, highlighting the importance of balancing, using ensemble techniques and statistical feature selection. Our models can potentially be used by clinicians after external validation.

Google Trends application for the study of information search behaviour on oropharyngeal cancer in Spain

PURPOSE

Oropharyngeal cancer is estimated to continue to increase in the next decades. Prevention strategies and knowing the current situation of knowledge and concern of the population about this disease is necessary. Infodemiology is valuable to monitor health information-seeking behaviour trends and epidemiology. The objective of this study is to analyze the use and evolution, through Google trends as a source of information, of internet-based information-seeking behaviour related to the oropharyngeal cancer in Spain and related to mass media stories.

METHODS

Using Google Trends, the terms "throat cancer', "HPV", "laryngeal cancer", "tonsil cancer" and "oral cancer". The searches volume and trend were analyzed using a Jointpoint regression method from January 2009 to July 2019.

RESULTS

The most searched term was "HPV", with a search volume index of 61, followed by "throat cancer" (SVI = 25). The trend of the term "HPV" increased 6.1% annually (p < 0.000), with a linear correlation of both terms of 0.52 (p < 0.000). The greatest number of searches was carried out in the north of Spain, the most repeated query being "oral sex AND cancer". A correlation between the news in the media and the increase in the volume of searches for the terms was found.

CONCLUSION

Any news stories, new interventions or aetiology related to oropharyngeal cancer can manifest as an increase in information-seeking behaviours for "throat cancer" on Google. Understanding healthcare information-seeking behaviour is essential in order to control and plan the quality of knowledge provided by health organisations, advocacy groups and health professionals regarding head and neck cancers.

Machine Learning-Based DNA Methylation Score for Fetal Exposure to Maternal Smoking: Development and Validation in Samples Collected from Adolescents and Adults

BACKGROUND

Fetal exposure to maternal smoking during pregnancy is associated with the development of noncommunicable diseases in the offspring. Maternal smoking may induce such long-term effects through persistent changes in the DNA methylome, which therefore hold the potential to be used as a biomarker of this early life exposure. With declining costs for measuring DNA methylation, we aimed to develop a DNA methylation score that can be used on adolescent DNA methylation data and thereby generate a score for in utero cigarette smoke exposure.

METHODS

We used machine learning methods to create a score reflecting exposure to maternal smoking during pregnancy. This score is based on peripheral blood measurements of DNA methylation (Illumina's Infinium HumanMethylation450K BeadChip). The score was developed and tested in the Raine Study with data from 995 white 17-y-old participants using 10-fold cross-validation. The score was further tested and validated in independent data from the Northern Finland Birth Cohort 1986 (NFBC1986) (16-y-olds) and 1966 (NFBC1966) (31-y-olds). Further, three previously proposed DNA methylation scores were applied for comparison. The final score was developed with 204 CpGs using elastic net regression.

RESULTS

Sensitivity and specificity values for the best performing previously developed classifier ("Reese Score") were 88% and 72% for Raine, 87% and 61% for NFBC1986 and 72% and 70% for NFBC1966, respectively; corresponding figures using the elastic net regression approach were 91% and 76% (Raine), 87% and 75% (NFBC1986), and 72% and 78% for NFBC1966.

CONCLUSION

We have developed a DNA methylation score for exposure to maternal smoking during pregnancy, outperforming the three previously developed scores. One possible application of the current score could be for model adjustment purposes or to assess its association with distal health outcomes where part of the effect can be attributed to maternal smoking. Further, it may provide a biomarker for fetal exposure to maternal smoking. https://doi.org/10.1289/EHP6076.

Classical and Deep Learning Paradigms for Detection and Validation of Key Genes of Risky Outcomes of HCV

Hepatitis C virus (HCV) is one of the most dangerous viruses worldwide. It is the foremost cause of the hepatic cirrhosis, and hepatocellular carcinoma, HCC. Detecting new key genes that play a role in the growth of HCC in HCV patients using machine learning techniques paves the way for producing accurate antivirals. In this work, there are two phases: detecting the up/downregulated genes using classical univariate and multivariate feature selection methods, and validating the retrieved list of genes using Insilico classifiers. However, the classification algorithms in the medical domain frequently suffer from a deficiency of training cases. Therefore, a deep neural network approach is proposed here to validate the significance of the retrieved genes in classifying the HCV-infected samples from the disinfected ones. The validation model is based on the artificial generation of new examples from the retrieved genes’ expressions using sparse autoencoders. Subsequently, the generated genes’ expressions data are used to train conventional classifiers. Our results in the first phase yielded a better retrieval of significant genes using Principal Component Analysis (PCA), a multivariate approach. The retrieved list of genes using PCA had a higher number of HCC biomarkers compared to the ones retrieved from the univariate methods. In the second phase, the classification accuracy can reveal the relevance of the extracted key genes in classifying the HCV-infected and disinfected samples.

PEnBayes: A Multi-Layered Ensemble Approach for Learning Bayesian Network Structure from Big Data

Discovering the Bayesian network (BN) structure from big datasets containing rich causal relationships is becoming increasingly valuable for modeling and reasoning under uncertainties in many areas with big data gathered from sensors due to high volume and fast veracity. Most of the current BN structure learning algorithms have shortcomings facing big data. First, learning a BN structure from the entire big dataset is an expensive task which often ends in failure due to memory constraints. Second, it is quite difficult to select a learner from numerous BN structure learning algorithms to consistently achieve good learning accuracy. Lastly, there is a lack of an intelligent method that merges separately learned BN structures into a well structured BN network. To address these shortcomings, we introduce a novel parallel learning approach called PEnBayes (Parallel Ensemble-based Bayesian network learning). PEnBayes starts with an adaptive data preprocessing phase that calculates the Appropriate Learning Size and intelligently divides a big dataset for fast distributed local structure learning. Then, PEnBayes learns a collection of local BN Structures in parallel using a two-layered weighted adjacent matrix-based structure ensemble method. Lastly, PEnBayes merges the local BN Structures into a global network structure using the structure ensemble method at the global layer. For the experiment, we generate big data sets by simulating sensor data from patient monitoring, transportation, and disease diagnosis domains. The Experimental results show that PEnBayes achieves a significantly improved execution performance with more consistent and stable results compared with three baseline learning algorithms.

Peripheral transcriptomic biomarkers for early detection of sporadic Alzheimer disease?

Alzheimer disease (AD) is the major epidemic of the 21^st century, its prevalence rising along with improved human longevity. Early AD diagnosis is key to successful treatment, as currently available therapeutics only allow small benefits for diagnosed AD patients. By contrast, future therapeutics, including those already in preclinical or clinical trials, are expected to afford neuroprotection prior to widespread brain damage and dementia. Brain imaging technologies are developing as promising tools for early AD diagnostics, yet their high cost limits their utility for screening at-risk populations. Blood or plasma transcriptomics, proteomics, and/or metabolomics may pave the way for cost-effective AD risk screening in middle-aged individuals years ahead of cognitive decline. This notion is exemplified by data mining of blood transcriptomics from a published dataset. Consortia blood sample collection and analysis from large cohorts with mild cognitive impairment followed longitudinally for their cognitive state would allow the development of a reliable and inexpensive early AD screening tool.

Big data analytics for preventive medicine

Medical data is one of the most rewarding and yet most complicated data to analyze. How can healthcare providers use modern data analytics tools and technologies to analyze and create value from complex data? Data analytics, with its promise to efficiently discover valuable pattern by analyzing large amount of unstructured, heterogeneous, non-standard and incomplete healthcare data. It does not only forecast but also helps in decision making and is increasingly noticed as breakthrough in ongoing advancement with the goal is to improve the quality of patient care and reduces the healthcare cost. The aim of this study is to provide a comprehensive and structured overview of extensive research on the advancement of data analytics methods for disease prevention. This review first introduces disease prevention and its challenges followed by traditional prevention methodologies. We summarize state-of-the-art data analytics algorithms used for classification of disease, clustering (unusually high incidence of a particular disease), anomalies detection (detection of disease) and association as well as their respective advantages, drawbacks and guidelines for selection of specific model followed by discussion on recent development and successful application of disease prevention methods. The article concludes with open research challenges and recommendations.

Peripheral transcriptomic biomarkers for early detection of sporadic Alzheimer disease?

Alzheimer disease (AD) is the major epidemic of the 21st century, its prevalence rising along with improved human longevity. Early AD diagnosis is key to successful treatment, as currently available therapeutics only allow small benefits for diagnosed AD patients. By contrast, future therapeutics, including those already in preclinical or clinical trials, are expected to afford neuroprotection prior to widespread brain damage and dementia. Brain imaging technologies are developing as promising tools for early AD diagnostics, yet their high cost limits their utility for screening at-risk populations. Blood or plasma transcriptomics, proteomics, and/or metabolomics may pave the way for cost-effective AD risk screening in middle-aged individuals years ahead of cognitive decline. This notion is exemplified by data mining of blood transcriptomics from a published dataset. Consortia blood sample collection and analysis from large cohorts with mild cognitive impairment followed longitudinally for their cognitive state would allow the development of a reliable and inexpensive early AD screening tool.

Causal Inference Network of Genes Related with Bone Metastasis of Breast Cancer and Osteoblasts Using Causal Bayesian Networks

Background

The causal networks among genes that are commonly expressed in osteoblasts and during bone metastasis (BM) of breast cancer (BC) are not well understood. Here, we developed a machine learning method to obtain a plausible causal network of genes that are commonly expressed during BM and in osteoblasts in BC.

Methods

We selected BC genes that are commonly expressed during BM and in osteoblasts from the Gene Expression Omnibus database. Bayesian Network Inference with Java Objects (Banjo) was used to obtain the Bayesian network. Genes registered as BC related genes were included as candidate genes in the implementation of Banjo. Next, we obtained the Bayesian structure and assessed the prediction rate for BM, conditional independence among nodes, and causality among nodes. Furthermore, we reported the maximum relative risks (RRs) of combined gene expression of the genes in the model.

Results

We mechanistically identified 33 significantly related and plausibly involved genes in the development of BC BM. Further model evaluations showed that 16 genes were enough for a model to be statistically significant in terms of maximum likelihood of the causal Bayesian networks (CBNs) and for correct prediction of BM of BC. Maximum RRs of combined gene expression patterns showed that the expression levels of UBIAD1, HEBP1, BTNL8, TSPO, PSAT1, and ZFP36L2 significantly affected development of BM from BC.

Conclusions

The CBN structure can be used as a reasonable inference network for accurately predicting BM in BC.

A Convenient Non-harm Cervical Spondylosis Intelligent Identity method based on Machine Learning

Cervical spondylosis (CS), a most common orthopedic diseases, is mainly identified by the doctor's judgment from the clinical symptoms and cervical change provided by expensive instruments in hospital. Owing to the development of the surface electromyography (sEMG) technique and artificial intelligence, we proposed a convenient non-harm CS intelligent identify method EasiCNCSII, including the sEMG data acquisition and the CS identification. Faced with the limit testable muscles, the data acquisition method are proposed to conveniently and effectively collect data based on the tendons theory and CS etiology. Faced with high-dimension and the weak availability of the data, the 3-tier model EasiAI is developed to intelligently identify CS. The common features and new features are extracted from raw sEMG data in first tier. The EasiRF is proposed in second tier to further reduce the data dimension, improving the performance. A classification model based on gradient boosted regression tree is developed in third tier to identify CS. Compared with 4 common machine learning classification models, the EasiCNCSII achieves best performance of 91.02% in mean accuracy, 97.14% in mean sensitivity, 81.43% in mean specificity, 0.95 in mean AUC.

Assessing the Methods, Tools, and Statistical Approaches in Google Trends Research: Systematic Review

Background

In the era of information overload, are big data analytics the answer to access and better manage available knowledge? Over the last decade, the use of Web-based data in public health issues, that is, infodemiology, has been proven useful in assessing various aspects of human behavior. Google Trends is the most popular tool to gather such information, and it has been used in several topics up to this point, with health and medicine being the most focused subject. Web-based behavior is monitored and analyzed in order to examine actual human behavior so as to predict, better assess, and even prevent health-related issues that constantly arise in everyday life.

Objective

This systematic review aimed at reporting and further presenting and analyzing the methods, tools, and statistical approaches for Google Trends (infodemiology) studies in health-related topics from 2006 to 2016 to provide an overview of the usefulness of said tool and be a point of reference for future research on the subject.

Methods

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for selecting studies, we searched for the term “Google Trends” in the Scopus and PubMed databases from 2006 to 2016, applying specific criteria for types of publications and topics. A total of 109 published papers were extracted, excluding duplicates and those that did not fall inside the topics of health and medicine or the selected article types. We then further categorized the published papers according to their methodological approach, namely, visualization, seasonality, correlations, forecasting, and modeling.

Results

All the examined papers comprised, by definition, time series analysis, and all but two included data visualization. A total of 23.1% (24/104) studies used Google Trends data for examining seasonality, while 39.4% (41/104) and 32.7% (34/104) of the studies used correlations and modeling, respectively. Only 8.7% (9/104) of the studies used Google Trends data for predictions and forecasting in health-related topics; therefore, it is evident that a gap exists in forecasting using Google Trends data.

Conclusions

The monitoring of online queries can provide insight into human behavior, as this field is significantly and continuously growing and will be proven more than valuable in the future for assessing behavioral changes and providing ground for research using data that could not have been accessed otherwise.

Modeling interrelationships between health behaviors in overweight breast cancer survivors: Applying Bayesian networks

Obesity and its impact on health is a multifaceted phenomenon encompassing many factors, including demographics, environment, lifestyle, and psychosocial functioning. A systems science approach, investigating these many influences, is needed to capture the complexity and multidimensionality of obesity prevention to improve health. Leveraging baseline data from a unique clinical cohort comprising 333 postmenopausal overweight or obese breast cancer survivors participating in a weight-loss trial, we applied Bayesian networks, a machine learning approach, to infer interrelationships between lifestyle factors (e.g., sleep, physical activity), body mass index (BMI), and health outcomes (biomarkers and self-reported quality of life metrics). We used bootstrap resampling to assess network stability and accuracy, and Bayesian information criteria (BIC) to compare networks. Our results identified important behavioral subnetworks. BMI was the primary pathway linking behavioral factors to glucose regulation and inflammatory markers; the BMI-biomarker link was reproduced in 100% of resampled networks. Sleep quality was a hub impacting mental quality of life and physical health with > 95% resampling reproducibility. Omission of the BMI or sleep links significantly degraded the fit of the networks. Our findings suggest potential mechanistic pathways and useful intervention targets for future trials. Using our models, we can make quantitative predictions about health impacts that would result from targeted, weight loss and/or sleep improvement interventions. Importantly, this work highlights the utility of Bayesian networks in health behaviors research.

Information-Based Medicine in Glioma Patients: A Clinical Perspective

Glioma constitutes the most common type of primary brain tumor with a dismal survival, often measured in terms of months or years. The thin line between treatment effectiveness and patient harm underpins the importance of tailoring clinical management to the individual patient. Randomized trials have laid the foundation for many neuro-oncological guidelines. Despite this, their findings focus on group-level estimates. Given our current tools, we are limited in our ability to guide patients on what therapy is best for them as individuals, or even how long they should expect to survive. Machine learning, however, promises to provide the analytical support for personalizing treatment decisions, and deep learning allows clinicians to unlock insight from the vast amount of unstructured data that is collected on glioma patients. Although these novel techniques have achieved astonishing results across a variety of clinical applications, significant hurdles remain associated with the implementation of them in clinical practice. Future challenges include the assembly of well-curated cross-institutional datasets, improvement of the interpretability of machine learning models, and balancing novel evidence-based decision-making with the associated liability of automated inference. Although artificial intelligence already exceeds clinical expertise in a variety of applications, clinicians remain responsible for interpreting the implications of, and acting upon, each prediction.

The revolution of personalized psychiatry: will technology make it happen sooner?

Personalized medicine (PM) aims to establish a new approach in clinical decision-making, based upon a patient's individual profile in order to tailor treatment to each patient's characteristics. Although this has become a focus of the discussion also in the psychiatric field, with evidence of its high potential coming from several proof-of-concept studies, nearly no tools have been developed by now that are ready to be applied in clinical practice. In this paper, we discuss recent technological advances that can make a shift toward a clinical application of the PM paradigm. We focus specifically on those technologies that allow both the collection of massive as much as real-time data, i.e., electronic medical records and smart wearable devices, and to achieve relevant predictions using these data, i.e. the application of machine learning techniques.

Principles of Experimental Design for Big Data Analysis

Big Datasets are endemic, but are often notoriously difficult to analyse because of their size, heterogeneity and quality. The purpose of this paper is to open a discourse on the potential for modern decision theoretic optimal experimental design methods, which by their very nature have traditionally been applied prospectively, to improve the analysis of Big Data through retrospective designed sampling in order to answer particular questions of interest. By appealing to a range of examples, it is suggested that this perspective on Big Data modelling and analysis has the potential for wide generality and advantageous inferential and computational properties. We highlight current hurdles and open research questions surrounding efficient computational optimisation in using retrospective designs, and in part this paper is a call to the optimisation and experimental design communities to work together in the field of Big Data analysis.

Analyzing Electronic Medical Records to Predict Risk of DIT (Death, Intubation, or Transfer to ICU) in Pediatric Respiratory Failure or Related Conditions

Large volumes of data are generated in hospital settings, including clinical and physiological data generated during the course of patient care. Our goal, as proof of concept, was to identify early clinical factors or traits useful for predicting the outcome, of death, intubation, or transfer to ICU, for children with pediatric respiratory failure. We implemented both supervised and unsupervised methods to extend our understanding on statistical relationships in clinical and physiological data. As a supervised learning method, we use binary logistic regression to predict the risk of developing DIT outcome. Next, we implemented unsupervised k-means algorithm on principal components of clinical and physiological data to further explore the contribution of clinical and physiological data on developing DIT outcome. Our results show that early signals of DIT can be detected in physiological data, and two risk factors, blood pressure and oxygen level, are the most important determinant of developing DIT.

Big and disparate data: considerations for pediatric consortia

PURPOSE OF REVIEW

Increasingly, there is a need for examining exposure disease associations in large, diverse datasets to understand the complex determinants of pediatric disease and disability. Recognizing that children's health research consortia will be important sources of big data, it is crucial for the pediatric research community to be knowledgeable about the challenges and opportunities that they will face. The present review will provide examples of existing children's health consortia, highlight recent pooled analyses conducted by children's health research consortia, address common challenges of pooled analyses, and provide recommendations to advance collective research efforts in pediatric research.

RECENT FINDINGS

Formal consortia and other collective-science initiatives are increasingly being created to share individual data from a set of relevant epidemiological studies to address a common research topic under the concept that the joint effort of many individual groups can accomplish far more than working alone. There are practical challenges to the participation of investigators within consortia that need to be addressed in order for them to work.

SUMMARY

Researchers who access consortia with data centers will be able to go far beyond their initial hypotheses and potentially accomplish research that was previously thought infeasible or too costly.

Machine learning approaches to personalize early prediction of asthma exacerbations

Patient telemonitoring results in an aggregation of significant amounts of information about patient disease trajectory. However, the potential use of this information for early prediction of exacerbations in adult asthma patients has not been systematically evaluated. The aim of this study was to explore the utility of telemonitoring data for building machine learning algorithms that predict asthma exacerbations before they occur. The study dataset comprised daily self-monitoring reports consisting of 7001 records submitted by adult asthma patients during home telemonitoring. Predictive modeling included preparation of stratified training datasets, predictive feature selection, and evaluation of resulting classifiers. Using a 7-day window, a naive Bayesian classifier, adaptive Bayesian network, and support vector machines were able to predict asthma exacerbation occurring on day 8, with sensitivity of 0.80, 1.00, and 0.84; specificity of 0.77, 1.00, and 0.80; and accuracy of 0.77, 1.00, and 0.80, respectively. Our study demonstrated that machine learning techniques have significant potential in developing personalized decision support for chronic disease telemonitoring systems. Future studies may benefit from a comprehensive predictive framework that combines telemonitoring data with other factors affecting the likelihood of developing acute exacerbation. Approaches implemented for advanced asthma exacerbation prediction may be extended to prediction of exacerbations in patients with other chronic health conditions.

Data, Big Data, and Metadata in Anesthesiology

The last decade has seen an explosion in the growth of digital data. Since 2005, the total amount of digital data created or replicated on all platforms and devices has been doubling every 2 years, from an estimated 132 exabytes (132 billion gigabytes) in 2005 to 4.4 zettabytes (4.4 trillion gigabytes) in 2013, and a projected 44 zettabytes (44 trillion gigabytes) in 2020. This growth has been driven in large part by the rise of social media along with more powerful and connected mobile devices, with an estimated 75% of information in the digital universe generated by individuals rather than entities. Transactions and communications including payments, instant messages, Web searches, social media updates, and online posts are all becoming part of a vast pool of data that live "in the cloud" on clusters of servers located in remote data centers. The amount of accumulating data has become so large that it has given rise to the term Big Data. In many ways, Big Data is just a buzzword, a phrase that is often misunderstood and misused to describe any sort of data, no matter the size or complexity. However, there is truth to the assertion that some data sets truly require new management and analysis techniques.

The symptom phenotype of oncology outpatients remains relatively stable from prior to through 1 week following chemotherapy

Some oncology outpatients experience a higher number of and more severe symptoms during chemotherapy (CTX). However, little is known about whether this high risk phenotype persists over time. Latent transition analysis (LTA) was used to examine the probability that patients remained in the same symptom class when assessed prior to the administration of and following their next dose of CTX. For the patients whose class membership remained consistent, differences in demographic and clinical characteristics, and quality of life (QOL) were evaluated. The Memorial Symptom Assessment Scale (MSAS) was used to evaluate symptom burden. LTA was used to identify subgroups of patients with distinct symptom experiences based on the occurrence of the MSAS symptoms. Of the 906 patients evaluated, 83.9% were classified in the same symptom occurrence class at both assessments. Of these 760 patients, 25.0% were classified as Low-Low, 44.1% as Moderate-Moderate and 30.9% as High-High. Compared to the Low-Low class, the other two classes were younger, more likely to be women and to report child care responsibilities, and had a lower functional status and a higher comorbidity scores. The two higher classes reported lower QOL scores. The use of LTA could assist clinicians to identify higher risk patients and initiate more aggressive interventions.

Progress in oral personalized medicine: contribution of 'omics'

BACKGROUND

Precision medicine (PM), representing clinically applicable personalized medicine, proactively integrates and interprets multidimensional personal health data, including clinical, 'omics', and environmental profiles, into clinical practice. Realization of PM remains in progress.

OBJECTIVE

The focus of this review is to provide a descriptive narrative overview of: 1) the current status of oral personalized medicine; and 2) recent advances in genomics and related 'omic' and emerging research domains contributing to advancing oral-systemic PM, with special emphasis on current understanding of oral microbiomes.

DESIGN

A scan of peer-reviewed literature describing oral PM or 'omic'-based research conducted on humans/data published in English within the last 5 years in journals indexed in the PubMed database was conducted using mesh search terms. An evidence-based approach was used to report on recent advances with potential to advance PM in the context of historical critical and systematic reviews to delineate current state-of-the-art technologies. Special focus was placed on oral microbiome research associated with health and disease states, emerging research domains, and technological advances, which are positioning realization of PM.

RESULTS

This review summarizes: 1) evolving conceptualization of personalized medicine; 2) emerging insight into roles of oral infectious and inflammatory processes as contributors to both oral and systemic diseases; 3) community shifts in microbiota that may contribute to disease; 4) evidence pointing to new uncharacterized potential oral pathogens; 5) advances in technological approaches to 'omics' research that will accelerate PM; 6) emerging research domains that expand insights into host-microbe interaction including inter-kingdom communication, systems and network analysis, and salivaomics; and 7) advances in informatics and big data analysis capabilities to facilitate interpretation of host and microbiome-associated datasets. Furthermore, progress in clinically applicable screening assays and biomarker definition to inform clinical care are briefly explored.

CONCLUSION

Advancement of oral PM currently remains in research and discovery phases. Although substantive progress has been made in advancing the understanding of the role of microbiome dynamics in health and disease and is being leveraged to advance early efforts at clinical translation, further research is required to discern interpretable constituency patterns in the complex interactions of these microbial communities in health and disease. Advances in biotechnology and bioinformatics facilitating novel approaches to rapid analysis and interpretation of large datasets are providing new insights into oral health and disease, potentiating clinical application and advancing realization of PM within the next decade.

Applying data envelopment analysis to preventive medicine: a novel method for constructing a personalized risk model of obesity

Data envelopment analysis (DEA) is a method of operations research that has not yet been applied in the field of obesity research. However, DEA might be used to evaluate individuals’ susceptibility to obesity, which could help establish effective risk models for the onset of obesity. Therefore, we conducted this study to evaluate the feasibility of applying DEA to predict obesity, by calculating efficiency scores and evaluating the usefulness of risk models. In this study, we evaluated data from the Takahata study, which was a population-based cohort study (with a follow-up study) of Japanese people who are >40 years old. For our analysis, we used the input-oriented Charnes-Cooper-Rhodes model of DEA, and defined the decision-making units (DMUs) as individual subjects. The inputs were defined as (1) exercise (measured as calories expended) and (2) the inverse of food intake (measured as calories ingested). The output was defined as the inverse of body mass index (BMI). Using the β coefficients for the participants’ single nucleotide polymorphisms, we then calculated their genetic predisposition score (GPS). Both efficiency scores and GPS were available for 1,620 participants from the baseline survey, and for 708 participants from the follow-up survey. To compare the strengths of the associations, we used models of multiple linear regressions. To evaluate the effects of genetic factors and efficiency score on body mass index (BMI), we used multiple linear regression analysis, with BMI as the dependent variable, GPS and efficiency scores as the explanatory variables, and several demographic controls, including age and sex. Our results indicated that all factors were statistically significant (p < 0.05), with an adjusted R² value of 0.66. Therefore, it is possible to use DEA to predict environmentally driven obesity, and thus to establish a well-fitted model for risk of obesity.

Data collection as a barrier to personalized medicine

Basic life science research holds the promise of personalizing medical care. However, translation steps from the laboratory to the bedside are not trivial. Results from clinical research are difficult to replicate in part because study cohorts are poorly defined phenotypically. Here, we discuss how computer technology can improve the collection of clinical data to enable translation of insights from basic science to validated clinical guidelines.