Big data and clinicians: a review on the state of the science

MHIF-MSEA: a novel model of miRNA set enrichment analysis based on multi-source heterogeneous information fusion

Introduction: MicroRNAs (miRNAs) are a class of non-coding RNA molecules that play a crucial role in the regulation of diverse biological processes across various organisms. Despite not encoding proteins, miRNAs have been found to have significant implications in the onset and progression of complex human diseases. Methods: Conventional methods for miRNA functional enrichment analysis have certain limitations, and we proposed a novel method called MiRNA Set Enrichment Analysis based on Multi-source Heterogeneous Information Fusion (MHIF-MSEA). Three miRNA similarity networks (miRSN-DA, miRSN-GOA, and miRSN-PPI) were constructed in MHIF-MSEA. These networks were built based on miRNA-disease association, gene ontology (GO) annotation of target genes, and protein-protein interaction of target genes, respectively. These miRNA similarity networks were fused into a single similarity network with the averaging method. This fused network served as the input for the random walk with restart algorithm, which expanded the original miRNA list. Finally, MHIF-MSEA performed enrichment analysis on the expanded list. Results and Discussion: To determine the optimal network fusion approach, three case studies were introduced: colon cancer, breast cancer, and hepatocellular carcinoma. The experimental results revealed that the miRNA-miRNA association network constructed using miRSN-DA and miRSN-GOA exhibited superior performance as the input network. Furthermore, the MHIF-MSEA model performed enrichment analysis on differentially expressed miRNAs in breast cancer and hepatocellular carcinoma. The achieved p-values were 2.17e(-75) and 1.50e(-77), and the hit rates improved by 39.01% and 44.68% compared to traditional enrichment analysis methods, respectively. These results confirm that the MHIF-MSEA method enhances the identification of enriched miRNA sets by leveraging multiple sources of heterogeneous information, leading to improved insights into the functional implications of miRNAs in complex diseases.

Advances in Machine Learning Processing of Big Data from Disease Diagnosis Sensors

Exploring accurate, noninvasive, and inexpensive disease diagnostic sensors is a critical task in the fields of chemistry, biology, and medicine. The complexity of biological systems and the explosive growth of biomarker data have driven machine learning to become a powerful tool for mining and processing big data from disease diagnosis sensors. With the development of bioinformatics and artificial intelligence (AI), machine learning models formed by data mining have been able to guide more sensitive and accurate molecular computing. This review presents an overview of big data collection approaches and fundamental machine learning algorithms and discusses recent advances in machine learning and molecular computational disease diagnostic sensors. More specifically, we highlight existing modular workflows and key opportunities and challenges for machine learning to achieve disease diagnosis through big data mining.

TLSEA: a tool for lncRNA set enrichment analysis based on multi-source heterogeneous information fusion

Long non-coding RNAs (lncRNAs) play an important regulatory role in gene transcription and post-transcriptional modification, and lncRNA regulatory dysfunction leads to a variety of complex human diseases. Hence, it might be beneficial to detect the underlying biological pathways and functional categories of genes that encode lncRNA. This can be carried out by using gene set enrichment analysis, which is a pervasive bioinformatic technique that has been widely used. However, accurately performing gene set enrichment analysis of lncRNAs remains a challenge. Most conventional enrichment analysis methods have not exhaustively included the rich association information among genes, which usually affects the regulatory functions of genes. Here, we developed a novel tool for lncRNA set enrichment analysis (TLSEA) to improve the accuracy of the gene functional enrichment analysis, which extracted the low-dimensional vectors of lncRNAs in two functional annotation networks with the graph representation learning method. A novel lncRNA-lncRNA association network was constructed by merging lncRNA-related heterogeneous information obtained from multiple sources with the different lncRNA-related similarity networks. In addition, the random walk with restart method was adopted to effectively expand the lncRNAs submitted by users according to the lncRNA-lncRNA association network of TLSEA. In addition, a case study of breast cancer was performed, which demonstrated that TLSEA could detect breast cancer more accurately than conventional tools. The TLSEA can be accessed freely at http://www.lirmed.com:5003/tlsea.

Improving child health through Big Data and data science

Child health is defined by a complex, dynamic network of genetic, cultural, nutritional, infectious, and environmental determinants at distinct, developmentally determined epochs from preconception to adolescence. This network shapes the future of children, susceptibilities to adult diseases, and individual child health outcomes. Evolution selects characteristics during fetal life, infancy, childhood, and adolescence that adapt to predictable and unpredictable exposures/stresses by creating alternative developmental phenotype trajectories. While child health has improved in the United States and globally over the past 30 years, continued improvement requires access to data that fully represent the complexity of these interactions and to new analytic methods. Big Data and innovative data science methods provide tools to integrate multiple data dimensions for description of best clinical, predictive, and preventive practices, for reducing racial disparities in child health outcomes, for inclusion of patient and family input in medical assessments, and for defining individual disease risk, mechanisms, and therapies. However, leveraging these resources will require new strategies that intentionally address institutional, ethical, regulatory, cultural, technical, and systemic barriers as well as developing partnerships with children and families from diverse backgrounds that acknowledge historical sources of mistrust. We highlight existing pediatric Big Data initiatives and identify areas of future research. IMPACT: Big Data and data science can improve child health. This review highlights the importance for child health of child-specific and life course-based Big Data and data science strategies. This review provides recommendations for future pediatric-specific Big Data and data science research.

Perspective Toward Machine Learning Implementation in Pediatric Medicine: Mixed Methods Study

Background

Given the costs of machine learning implementation, a systematic approach to prioritizing which models to implement into clinical practice may be valuable.

Objective

The primary objective was to determine the health care attributes respondents at 2 pediatric institutions rate as important when prioritizing machine learning model implementation. The secondary objective was to describe their perspectives on implementation using a qualitative approach.

Methods

In this mixed methods study, we distributed a survey to health system leaders, physicians, and data scientists at 2 pediatric institutions. We asked respondents to rank the following 5 attributes in terms of implementation usefulness: the clinical problem was common, the clinical problem caused substantial morbidity and mortality, risk stratification led to different actions that could reasonably improve patient outcomes, reducing physician workload, and saving money. Important attributes were those ranked as first or second most important. Individual qualitative interviews were conducted with a subsample of respondents.

Results

Among 613 eligible respondents, 275 (44.9%) responded. Qualitative interviews were conducted with 17 respondents. The most common important attributes were risk stratification leading to different actions (205/275, 74.5%) and clinical problem causing substantial morbidity or mortality (177/275, 64.4%). The attributes considered least important were reducing physician workload and saving money. Qualitative interviews consistently prioritized implementations that improved patient outcomes.

Conclusions

Respondents prioritized machine learning model implementation where risk stratification would lead to different actions and clinical problems that caused substantial morbidity and mortality. Implementations that improved patient outcomes were prioritized. These results can help provide a framework for machine learning model implementation.

U.S. government, health-related data sources accessible to health researchers: A mapping review (Preprint)

BACKGROUND

Big data from large, government-sponsored surveys and data sets offers researchers opportunities to conduct population-based studies of important health issues in the United States, as well as develop preliminary data to support proposed future work. Yet, navigating these national data sources is challenging. Despite the widespread availability of national data, there is little guidance for researchers on how to access and evaluate the use of these resources.

OBJECTIVE

Our aim was to identify and summarize a comprehensive list of federally sponsored, health- and health care-related data sources that are accessible in the public domain in order to facilitate their use by researchers.

METHODS

We conducted a systematic mapping review of government sources of health-related data on US populations and with active or recent (previous 10 years) data collection. The key measures were government sponsor, overview and purpose of data, population of interest, sampling design, sample size, data collection methodology, type and description of data, and cost to obtain data. Convergent synthesis was used to aggregate findings.

RESULTS

Among 106 unique data sources, 57 met the inclusion criteria. Data sources were classified as survey or assessment data (n=30, 53%), trends data (n=27, 47%), summative processed data (n=27, 47%), primary registry data (n=17, 30%), and evaluative data (n=11, 19%). Most (n=39, 68%) served more than 1 purpose. The population of interest included individuals/patients (n=40, 70%), providers (n=15, 26%), and health care sites and systems (n=14, 25%). The sources collected data on demographic (n=44, 77%) and clinical information (n=35, 61%), health behaviors (n=24, 42%), provider or practice characteristics (n=22, 39%), health care costs (n=17, 30%), and laboratory tests (n=8, 14%). Most (n=43, 75%) offered free data sets.

CONCLUSIONS

A broad scope of national health data is accessible to researchers. These data provide insights into important health issues and the nation's health care system while eliminating the burden of primary data collection. Data standardization and uniformity were uncommon across government entities, highlighting a need to improve data consistency. Secondary analyses of national data are a feasible, cost-efficient means to address national health concerns.

Integrated multimodal artificial intelligence framework for healthcare applications

Artificial intelligence (AI) systems hold great promise to improve healthcare over the next decades. Specifically, AI systems leveraging multiple data sources and input modalities are poised to become a viable method to deliver more accurate results and deployable pipelines across a wide range of applications. In this work, we propose and evaluate a unified Holistic AI in Medicine (HAIM) framework to facilitate the generation and testing of AI systems that leverage multimodal inputs. Our approach uses generalizable data pre-processing and machine learning modeling stages that can be readily adapted for research and deployment in healthcare environments. We evaluate our HAIM framework by training and characterizing 14,324 independent models based on HAIM-MIMIC-MM, a multimodal clinical database (N = 34,537 samples) containing 7279 unique hospitalizations and 6485 patients, spanning all possible input combinations of 4 data modalities (i.e., tabular, time-series, text, and images), 11 unique data sources and 12 predictive tasks. We show that this framework can consistently and robustly produce models that outperform similar single-source approaches across various healthcare demonstrations (by 6–33%), including 10 distinct chest pathology diagnoses, along with length-of-stay and 48 h mortality predictions. We also quantify the contribution of each modality and data source using Shapley values, which demonstrates the heterogeneity in data modality importance and the necessity of multimodal inputs across different healthcare-relevant tasks. The generalizable properties and flexibility of our Holistic AI in Medicine (HAIM) framework could offer a promising pathway for future multimodal predictive systems in clinical and operational healthcare settings.

Characterization of Traumatic Injury During the Early COVID-19 Pandemic: Results From a National Healthcare Database

OBJECTIVE

To characterize traumatic injury patterns after stay-at-home orders were implemented in the United States in response to the coronavirus disease 2019 (COVID-19).

METHODS

A retrospective review of a convenience sample of patients from a national healthcare research database (TriNetX) was conducted from April 1, 2020, to June 30, 2020. Inclusion criteria included all patients with documentation of both injury pattern and mechanism of injury. A comparison was made to a matched pre-pandemic timeframe. Changes in percentage and rate ratio (RR) with a 95% confidence interval were reported. RRs were calculated using Poisson regression analysis.

RESULTS

 A total of 238,661 patients in the control and 178,224 patients in the study cohorts were analyzed. Significant increases in assaults (RR: 1.17, 95% CI: 1.14, 1.20) and bicycle accidents (RR: 1.07, 95% CI: 1.04, 1.11) were noted. There was a relative increase in patients who were male (+1.78%) and white (+2.01%). More injuries were alcohol-related (+0.76%) and occurred at home (+0.79%). A decrease in motor vehicle accidents (-1.17%), foot and ankle injuries (-1.63%), and injuries occurring at sporting events (-0.54%) was noted.

CONCLUSIONS

Changes in injury patterns were observed during the study period. During future crises, particular public health and injury prevention resources may be required to address assaults, substance abuse, and home safety.

Construction of a Big Data Platform in Healthcare with Multi-source, Heterogeneous Data Integration and Massive High-Dimensional Data Governance for Large Hospitals: Design, Development, and Application (Preprint)

Background

With the advent of data-intensive science, a full integration of big data science and health care will bring a cross-field revolution to the medical community in China. The concept big data represents not only a technology but also a resource and a method. Big data are regarded as an important strategic resource both at the national level and at the medical institutional level, thus great importance has been attached to the construction of a big data platform for health care.

Objective

We aimed to develop and implement a big data platform for a large hospital, to overcome difficulties in integrating, calculating, storing, and governing multisource heterogeneous data in a standardized way, as well as to ensure health care data security.

Methods

The project to build a big data platform at West China Hospital of Sichuan University was launched in 2017. The West China Hospital of Sichuan University big data platform has extracted, integrated, and governed data from different departments and sections of the hospital since January 2008. A master–slave mode was implemented to realize the real-time integration of multisource heterogeneous massive data, and an environment that separates heterogeneous characteristic data storage and calculation processes was built. A business-based metadata model was improved for data quality control, and a standardized health care data governance system and scientific closed-loop data security ecology were established.

Results

After 3 years of design, development, and testing, the West China Hospital of Sichuan University big data platform was formally brought online in November 2020. It has formed a massive multidimensional data resource database, with more than 12.49 million patients, 75.67 million visits, and 8475 data variables. Along with hospital operations data, newly generated data are entered into the platform in real time. Since its launch, the platform has supported more than 20 major projects and provided data service, storage, and computing power support to many scientific teams, facilitating a shift in the data support model—from conventional manual extraction to self-service retrieval (which has reached 8561 retrievals per month).

Conclusions

The platform can combine operation systems data from all departments and sections in a hospital to form a massive high-dimensional high-quality health care database that allows electronic medical records to be used effectively and taps into the value of data to fully support clinical services, scientific research, and operations management. The West China Hospital of Sichuan University big data platform can successfully generate multisource heterogeneous data storage and computing power. By effectively governing massive multidimensional data gathered from multiple sources, the West China Hospital of Sichuan University big data platform provides highly available data assets and thus has a high application value in the health care field. The West China Hospital of Sichuan University big data platform facilitates simpler and more efficient utilization of electronic medical record data for real-world research.

Big data in the healthcare system: a synergy with artificial intelligence and blockchain technology

In the last decades big data has facilitating and improving our daily duties in the medical research and clinical fields; the strategy to get to this point is understanding how to organize and analyze the data in order to accomplish the final goal that is improving healthcare system, in terms of cost and benefits, quality of life and outcome patient. The main objective of this review is to illustrate the state-of-art of big data in healthcare, its features and architecture. We also would like to demonstrate the different application and principal mechanisms of big data in the latest technologies known as blockchain and artificial intelligence, recognizing their benefits and limitations. Perhaps, medical education and digital anatomy are unexplored fields that might be profitable to investigate as we are proposing. The healthcare system can be revolutionized using these different technologies. Thus, we are explaining the basis of these systems focused to the medical arena in order to encourage medical doctors, nurses, biotechnologies and other healthcare professions to be involved and create a more efficient and efficacy system.

The social licence for data-intensive health research: towards co-creation, public value and trust

Background

The rise of Big Data-driven health research challenges the assumed contribution of medical research to the public good, raising questions about whether the status of such research as a common good should be taken for granted, and how public trust can be preserved. Scandals arising out of sharing data during medical research have pointed out that going beyond the requirements of law may be necessary for sustaining trust in data-intensive health research. We propose building upon the use of a social licence for achieving such ethical governance.

Main text

We performed a narrative review of the social licence as presented in the biomedical literature. We used a systematic search and selection process, followed by a critical conceptual analysis. The systematic search resulted in nine publications. Our conceptual analysis aims to clarify how societal permission can be granted to health research projects which rely upon the reuse and/or linkage of health data. These activities may be morally demanding. For these types of activities, a moral legitimation, beyond the limits of law, may need to be sought in order to preserve trust. Our analysis indicates that a social licence encourages us to recognise a broad range of stakeholder interests and perspectives in data-intensive health research. This is especially true for patients contributing data. Incorporating such a practice paves the way towards an ethical governance, based upon trust. Public engagement that involves patients from the start is called for to strengthen this social licence.

Conclusions

There are several merits to using the concept of social licence as a guideline for ethical governance. Firstly, it fits the novel scale of data-related risks; secondly, it focuses attention on trustworthiness; and finally, it offers co-creation as a way forward. Greater trust can be achieved in the governance of data-intensive health research by highlighting strategic dialogue with both patients contributing the data, and the public in general. This should ultimately contribute to a more ethical practice of governance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12910-021-00677-5.

Data-Driven Surveillance: Effective Collection, Integration, and Interpretation of Data to Support Decision Making

The biggest change brought about by the “era of big data” to health in general, and epidemiology in particular, relates arguably not to the volume of data encountered, but to its variety. An increasing number of new data sources, including many not originally collected for health purposes, are now being used for epidemiological inference and contextualization. Combining evidence from multiple data sources presents significant challenges, but discussions around this subject often confuse issues of data access and privacy, with the actual technical challenges of data integration and interoperability. We review some of the opportunities for connecting data, generating information, and supporting decision-making across the increasingly complex “variety” dimension of data in population health, to enable data-driven surveillance to go beyond simple signal detection and support an expanded set of surveillance goals.

Microbial natural product databases: moving forward in the multi-omics era

Covering: 2010-2020The digital revolution is driving significant changes in how people store, distribute, and use information. With the advent of new technologies around linked data, machine learning and large-scale network inference, the natural products research field is beginning to embrace real-time sharing and large-scale analysis of digitized experimental data. Databases play a key role in this, as they allow systematic annotation and storage of data for both basic and advanced applications. The quality of the content, structure, and accessibility of these databases all contribute to their usefulness for the scientific community in practice. This review covers the development of databases relevant for microbial natural product discovery during the past decade (2010-2020), including repositories of chemical structures/properties, metabolomics, and genomic data (biosynthetic gene clusters). It provides an overview of the most important databases and their functionalities, highlights some early meta-analyses using such databases, and discusses basic principles to enable widespread interoperability between databases. Furthermore, it points out conceptual and practical challenges in the curation and usage of natural products databases. Finally, the review closes with a discussion of key action points required for the field moving forward, not only for database developers but for any scientist active in the field.

Development and utility assessment of a machine learning bloodstream infection classifier in pediatric patients receiving cancer treatments

Background

Objectives were to build a machine learning algorithm to identify bloodstream infection (BSI) among pediatric patients with cancer and hematopoietic stem cell transplantation (HSCT) recipients, and to compare this approach with presence of neutropenia to identify BSI.

Methods

We included patients 0–18 years of age at cancer diagnosis or HSCT between January 2009 and November 2018. Eligible blood cultures were those with no previous blood culture (regardless of result) within 7 days. The primary outcome was BSI. Four machine learning algorithms were used: elastic net, support vector machine and two implementations of gradient boosting machine (GBM and XGBoost). Model training and evaluation were performed using temporally disjoint training (60%), validation (20%) and test (20%) sets. The best model was compared to neutropenia alone in the test set.

Results

Of 11,183 eligible blood cultures, 624 (5.6%) were positive. The best model in the validation set was GBM, which achieved an area-under-the-receiver-operator-curve (AUROC) of 0.74 in the test set. Among the 2236 in the test set, the number of false positives and specificity of GBM vs. neutropenia were 508 vs. 592 and 0.76 vs. 0.72 respectively. Among 139 test set BSIs, six (4.3%) non-neutropenic patients were identified by GBM. All received antibiotics prior to culture result availability.

Conclusions

We developed a machine learning algorithm to classify BSI. GBM achieved an AUROC of 0.74 and identified 4.3% additional true cases in the test set. The machine learning algorithm did not perform substantially better than using presence of neutropenia alone to predict BSI.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-020-07618-2.

Facilitating the Development of Deep Learning Models with Visual Analytics for Electronic Health Records

Electronic health record (EHR) data are widely used to perform early diagnoses and create treatment plans, which are key areas of research. We aimed to increase the efficiency of iteratively applying data-intensive technology and verifying the results for complex and big EHR data. We used a system entailing sequence mining, interpretable deep learning models, and visualization on data extracted from the MIMIC-IIIdatabase for a group of patients diagnosed with heart disease. The results of sequence mining corresponded to specific pathways of interest to medical staff and were used to select patient groups that underwent these pathways. An interactive Sankey diagram representing these pathways and a heat map visually representing the weight of each variable were developed for temporal and quantitative illustration. We applied the proposed system to predict unplanned cardiac surgery using clinical pathways determined by sequence pattern mining to select cardiac surgery from complex EHRs to label subject groups and deep learning models. The proposed system aids in the selection of pathway-based patient groups, simplification of labeling, and exploratory the interpretation of the modeling results. The proposed system can help medical staff explore various pathways that patients have undergone and further facilitate the testing of various clinical hypotheses using big data in the medical domain.

Big Data Applications in Medical Field: A Literature Review

Digital imaging and medical reporting have acquired an essential role in healthcare, but the main challenge is the storage of a high volume of patient data. Although newer technologies are already introduced in the medical sciences to save records size, Big Data provides advancements by storing a large amount of data to improve the efficiency and quality of patient treatment with better care. It provides intelligent automation capabilities to reduce errors than manual inputs. Large numbers of research papers on big data in the medical field are studied and analyzed for their impacts, benefits, and applications. Big data has great potential to support the digitalization of all medical and clinical records and then save the entire data regarding the medical history of an individual or a group. This paper discusses big data usage for various industries and sectors. Finally, 12 significant applications for the medical field by the implementation of big data are identified and studied with a brief description. This technology can be gainfully used to extract useful information from the available data by analyzing and managing them through a combination of hardware and software. With technological advancement, big data provides health-related information for millions of patient-related to life issues such as lab tests reporting, clinical narratives, demographics, prescription, medical diagnosis, and related documentation. Thus, Big Data is essential in developing a better yet efficient analysis and storage healthcare services. The demand for big data applications is increasing due to its capability of handling and analyzing massive data. Not only in the future but even now, Big Data is proving itself as an axiom of storing, developing, analyzing, and providing overall health information to the physicians.

Development of an Innovative Tool to Appraise Big Data for Best Evidence

BACKGROUND

Datum from electronic sources has accumulated and resulted in the establishment of big data and data science. Big data consists of data sets that are larger than traditional data processing applications can manage. Data science is the research method used to analyze big data. Researchers are applying research methods to harness large and complex data sets to increase our understanding of population health by creating predictive models of patients using a variety of key variables or characteristics. Evidence-based practice relies on the appraisal of research to ensure rigor prior to implementation in clinical settings. Consistent with other research methods, papers based on data science should be subject to appraisal for determination of best evidence. The purpose of this paper is to present a tool that can be used to appraise research papers based on large data sets and data science research methods.

METHODS

The following approach was used to develop the Data Science Appraisal Tool (DSAT). Despite an exhaustive search, we were unable to locate an appraisal tool for papers based on data science research methods. We then synthesized the extant literature to form the tool. The tool is based on the common characteristics of big data: (a) verification that the data set is representative of big data; (b) preparation of the data for analysis; (c) methodology used for data analysis; (d) results; and (e) theoretically based.

LINKING EVIDENCE TO ACTION

Appraisal tools currently exist for traditional and well-known research methods. The DSAT provides a method to appraise papers based in data science for best evidence.

Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare

Precision Medicine implies a deep understanding of inter-individual differences in health and disease that are due to genetic and environmental factors. To acquire such understanding there is a need for the implementation of different types of technologies based on artificial intelligence (AI) that enable the identification of biomedically relevant patterns, facilitating progress towards individually tailored preventative and therapeutic interventions. Despite the significant scientific advances achieved so far, most of the currently used biomedical AI technologies do not account for bias detection. Furthermore, the design of the majority of algorithms ignore the sex and gender dimension and its contribution to health and disease differences among individuals. Failure in accounting for these differences will generate sub-optimal results and produce mistakes as well as discriminatory outcomes. In this review we examine the current sex and gender gaps in a subset of biomedical technologies used in relation to Precision Medicine. In addition, we provide recommendations to optimize their utilization to improve the global health and disease landscape and decrease inequalities.

Clinical Pharmacokinetic Studies in Pregnant Women and the Relevance of Pharmacometric Tools

BACKGROUND

In clinical pharmacokinetic (PK) studies, pregnant women are significantly underrepresented because of ethical and legal reasons which lead to a paucity of information on potential PK changes in this population. As a consequence, pharmacometric tools became instrumental to explore and quantify the impact of PK changes during pregnancy.

METHODS

We explore and discuss the typical characteristics of population PK and physiologically based pharmacokinetic (PBPK) models with a specific focus on pregnancy and postpartum.

RESULTS

Population PK models enable the analysis of dense, sparse or unbalanced data to explore covariates in order to (partly) explain inter-individual variability (including pregnancy) and to individualize dosing. For population PK models, we subsequently used an illustrative approach with ketorolac data to highlight the relevance of enantiomer specific modeling for racemic drugs during pregnancy, while data on antibiotic prophylaxis (cefazolin) during surgery illustrate the specific characteristics of the fetal compartments in the presence of timeconcentration profiles. For PBPK models, an overview on the current status of reports and papers during pregnancy is followed by a PBPK cefuroxime model to illustrate the added benefit of PBPK in evaluating dosing regimens in pregnant women.

CONCLUSIONS

Pharmacometric tools became very instrumental to improve perinatal pharmacology. However, to reach their full potential, multidisciplinary collaboration and structured efforts are needed to generate more information from already available datasets, to share data and models, and to stimulate cross talk between clinicians and pharmacometricians to generate specific observations (pathophysiology during pregnancy, breastfeeding) needed to further develop the field.

Brief introduction of medical database and data mining technology in big data era

Data mining technology can search for potentially valuable knowledge from a large amount of data, mainly divided into data preparation and data mining, and expression and analysis of results. It is a mature information processing technology and applies database technology. Database technology is a software science that researches manages, and applies databases. The data in the database are processed and analyzed by studying the underlying theory and implementation methods of the structure, storage, design, management, and application of the database. We have introduced several databases and data mining techniques to help a wide range of clinical researchers better understand and apply database technology.

Your data is gold – Data donation for better healthcare?

Today, medical data such as diagnoses, procedures, imaging reports and laboratory tests, are not only collected in context of primary research and clinical studies. In addition, citizens are tracking their daily steps, food intake, sport exercises, and disease symptoms via mobile phones and wearable devices. In this context, the topic of “data donation” is drawing increased attention in science, politics, ethics and practice. This paper provides insights into the status quo of personal data donation in Germany and from a global perspective. As this topic requires a consideration of several perspectives, potential benefits and related, multifaceted challenges for citizens, patients and researchers are discussed. This includes aspects such as data quality & accessibility, privacy and ethical considerations.

Exploring Big Data Analytic Approaches to Cancer Blog Text Analysis

In this article, the authors explore the potential of a big data analytics approach to unstructured text analytics of cancer blogs. The application is developed using Cloudera platform's Hadoop MapReduce framework. It uses several text analytics algorithms, including word count, word association, clustering, and classification, to identify and analyze the patterns and keywords in cancer blog postings. This article establishes an exploratory approach to involving big data analytics methods in developing text analytics applications for the analysis of cancer blogs. Additional insights are extracted through various means, including the development of categories or keywords contained in the blogs, the development of a taxonomy, and the examination of relationships among the categories. The application has the potential for generalizability and implementation with health content in other blogs and social media. It can provide insight and decision support for cancer management and facilitate efficient and relevant searches for information related to cancer.

Developing routinely recorded clinical data from electronic patient records as a national resource to improve neonatal health care: the Medicines for Neonates research programme

Background

Clinical data offer the potential to advance patient care. Neonatal specialised care is a high-cost NHS service received by approximately 80,000 newborn infants each year.

Objectives

(1) To develop the use of routinely recorded operational clinical data from electronic patient records (EPRs), secure national coverage, evaluate and improve the quality of clinical data, and develop their use as a national resource to improve neonatal health care and outcomes. To test the hypotheses that (2) clinical and research data are of comparable quality, (3) routine NHS clinical assessment at the age of 2 years reliably identifies children with neurodevelopmental impairment and (4) trial-based economic evaluations of neonatal interventions can be reliably conducted using clinical data. (5) To test methods to link NHS data sets and (6) to evaluate parent views of personal data in research.

Design

Six inter-related workstreams; quarterly extractions of predefined data from neonatal EPRs; and approvals from the National Research Ethics Service, Health Research Authority Confidentiality Advisory Group, Caldicott Guardians and lead neonatal clinicians of participating NHS trusts.

Setting

NHS neonatal units.

Participants

Neonatal clinical teams; parents of babies admitted to NHS neonatal units.

Interventions

In workstream 3, we employed the Bayley-III scales to evaluate neurodevelopmental status and the Quantitative Checklist of Autism in Toddlers (Q-CHAT) to evaluate social communication skills. In workstream 6, we recruited parents with previous experience of a child in neonatal care to assist in the design of a questionnaire directed at the parents of infants admitted to neonatal units.

Data sources

Data were extracted from the EPR of admissions to NHS neonatal units.

Main outcome measures

We created a National Neonatal Research Database (NNRD) containing a defined extract from real-time, point-of-care, clinician-entered EPRs from all NHS neonatal units in England, Wales and Scotland (n = 200), established a UK Neonatal Collaborative of all NHS trusts providing neonatal specialised care, and created a new NHS information standard: the Neonatal Data Set (ISB 1595) (see http://webarchive.nationalarchives.gov.uk/±/http://www.isb.nhs.uk/documents/isb-1595/amd-32–2012/index_html; accessed 25 June 2018).

Results

We found low discordance between clinical (NNRD) and research data for most important infant and maternal characteristics, and higher prevalence of clinical outcomes. Compared with research assessments, NHS clinical assessment at the age of 2 years has lower sensitivity but higher specificity for identifying children with neurodevelopmental impairment. Completeness and quality are higher for clinical than for administrative NHS data; linkage is feasible and substantially enhances data quality and scope. The majority of hospital resource inputs for economic evaluations of neonatal interventions can be extracted reliably from the NNRD. In general, there is strong parent support for sharing routine clinical data for research purposes.

Limitations

We were only able to include data from all English neonatal units from 2012 onwards and conduct only limited cross validation of NNRD data directly against data in paper case notes. We were unable to conduct qualitative analyses of parent perspectives. We were also only able to assess the utility of trial-based economic evaluations of neonatal interventions using a single trial. We suggest that results should be validated against other trials.

Conclusions

We show that it is possible to obtain research-standard data from neonatal EPRs, and achieve complete population coverage, but we highlight the importance of implementing systematic examination of NHS data quality and completeness and testing methods to improve these measures. Currently available EPR data do not enable ascertainment of neurodevelopmental outcomes reliably in very preterm infants. Measures to maintain high quality and completeness of clinical and administrative data are important health service goals. As parent support for sharing clinical data for research is underpinned by strong altruistic motivation, improving wider public understanding of benefits may enhance informed decision-making.

Future work

We aim to implement a new paradigm for newborn health care in which continuous incremental improvement is achieved efficiently and cost-effectively by close integration of evidence generation with clinical care through the use of high-quality EPR data. In future work, we aim to automate completeness and quality checks and make recording processes more ‘user friendly’ and constructed in ways that minimise the likelihood of missing or erroneous entries. The development of criteria that provide assurance that data conform to prespecified completeness and quality criteria would be an important development. The benefits of EPR data might be extended by testing their use in large pragmatic clinical trials. It would also be of value to develop methods to quality assure EPR data including involving parents, and link the NNRD to other health, social care and educational data sets to facilitate the acquisition of lifelong outcomes across multiple domains.

Study registration

This study is registered as PROSPERO CRD42015017439 (workstream 1) and PROSPERO CRD42012002168 (workstream 3).

Funding

The National Institute for Health Research Programme Grants for Applied Research programme (£1,641,471). Unrestricted donations were supplied by Abbott Laboratories (Maidenhead, UK: £35,000), Nutricia Research Foundation (Schiphol, the Netherlands: £15,000), GE Healthcare (Amersham, UK: £1000). A grant to support the use of routinely collected, standardised, electronic clinical data for audit, management and multidisciplinary feedback in neonatal medicine was received from the Department of Health and Social Care (£135,494).

Chronic low back pain is highly individualised: patterns of classification across three unidimensional subgrouping analyses

BACKGROUND AND AIMS

Chronic low back pain (CLBP) is a complex disorder where central and peripheral nociceptive processes are influenced by factors from multiple dimensions associated with CLBP (e.g. movement, pain sensitivity, psychological). To date, outcomes for treatments matched to unidimensional subgroups (e.g. psychologically-based) have been poor. Therefore, unidimensional subgrouping may not reflect the complexity of CLBP presentations at an individual level. The aim of this study was therefore to explore patterns of classification at an individual level across the three previously-published, data-driven, within-dimension subgrouping studies.

METHODS

Cross-sectional, multidimensional data was collected in 294 people with CLBP. Statistical derivation of subgroups within each of three clinically-important dimensions (pain sensitivity, psychological profile, pain responses following repeated spinal bending) was briefly reviewed. Patterns of classification membership were subsequently tabulated across the three dimensions.

RESULTS

Of 27 possible patterns across these dimensions, 26 were represented across the cohort.

CONCLUSIONS

This result highlights that while unidimensional subgrouping has been thought useful to guide treatment, it is unlikely to capture the full complexity of CLBP. The amount of complexity important for best patient outcomes is currently untested.

IMPLICATIONS

For clinicians this study highlights the high variability of presentations of people with CLBP at the level of the individual. For example, clinician's should not assume that those with high levels of pain sensitivity will also have high psychological distress and have pain summation following repeated spinal bending. A more flexible, multidimensional, clinically-reasoned approach to profile patient complexity may be required to inform individualised, patient-centred care. Such individualised care might improve treatment efficacy. This study also has implications for researchers; highlighting the inadequacy of unidimensional subgrouping processes and methodological difficulties in deriving subgroups across multidimensional data.

Big Health Data and Cardiovascular Diseases: A Challenge for Research, an Opportunity for Clinical Care

Cardiovascular disease (CVD) accounts for the majority of death and hospitalization, health care expenditures and loss of productivity in developed country. CVD research, thus, plays a key role for improving patients' outcomes as well as for the sustainability of health systems. The increasing costs and complexity of modern medicine along with the fragmentation in healthcare organizations interfere with improving quality care and represent a missed opportunity for research. The advancement in diagnosis, therapy and prognostic evaluation of patients with CVD, indeed, is frustrated by limited data access to selected small patient populations, not standardized nor computable definition of disease and lack of approved relevant patient-centered outcomes. These critical issues results in a deep mismatch between randomized controlled trials and real-world setting, heterogeneity in treatment response and wide inter-individual variation in prognosis. Big data approach combines millions of people's electronic health records (EHR) from different resources and provides a new methodology expanding data collection in three direction: high volume, wide variety and extreme acquisition speed. Large population studies based on EHR holds much promise due to low costs, diminished study participant burden, and reduced selection bias, thus offering an alternative to traditional ascertainment through biomedical screening and tracing processes. By merging and harmonizing large data sets, the researchers aspire to build algorithms that allow targeted and personalized CVD treatments. In current paper, we provide a critical review of big health data for cardiovascular research, focusing on the opportunities of this largely free data analytics and the challenges in its realization.

Personalized Approach and Precision Medicine in Supportive and End-of-Life Care for Patients With Advanced and End-Stage Kidney Disease

Kidney supportive care requires a highly personalized approach to care. Precision medicine holds promise for a deeper understanding of the pathophysiology of symptoms and related syndromes and more precise individualization of prognosis and treatment estimates, therefore providing valuable opportunities for greater personalization of supportive care. However, the major drivers of quality of life are psychosocial, economic, lifestyle, and preference-based, and consideration of these factors and skilled communication are integral to the provision of excellent and personalized kidney supportive care. This article discusses the concepts of personalized and precision medicine in the context of kidney supportive care and highlights some opportunities and limitations within these fields.

Data Analysis Strategies in Medical Imaging

Radiographic imaging continues to be one of the most effective and clinically useful tools within oncology. Sophistication of artificial intelligence has allowed for detailed quantification of radiographic characteristics of tissues using predefined engineered algorithms or deep learning methods. Precedents in radiology as well as a wealth of research studies hint at the clinical relevance of these characteristics. However, critical challenges are associated with the analysis of medical imaging data. Although some of these challenges are specific to the imaging field, many others like reproducibility and batch effects are generic and have already been addressed in other quantitative fields such as genomics. Here, we identify these pitfalls and provide recommendations for analysis strategies of medical imaging data, including data normalization, development of robust models, and rigorous statistical analyses. Adhering to these recommendations will not only improve analysis quality but also enhance precision medicine by allowing better integration of imaging data with other biomedical data sources. Clin Cancer Res; 24(15); 3492-9. ©2018 AACR.

Models Used in Clinical Decision Support Systems Supporting Healthcare Professionals Treating Chronic Wounds: Systematic Literature Review

Background

Chronic wounds such as diabetic foot ulcers, venous leg ulcers, and pressure ulcers are a massive burden to health care facilities. Many randomized controlled trials on different wound care elements have been conducted and published in the Cochrane Library, all of which have only a low evidential basis. Thus, health care professionals are forced to rely on their own experience when making decisions regarding wound care. To progress from experience-based practice to evidence-based wound care practice, clinical decision support systems (CDSS) that help health care providers with decision-making in a clinical workflow have been developed. These systems have proven useful in many areas of the health care sector, partly because they have increased the quality of care, and partially because they have generated a solid basis for evidence-based practice. However, no systematic reviews focus on CDSS within the field of wound care to chronic wounds.

Objective

The aims of this systematic literature review are (1) to identify models used in CDSS that support health care professionals treating chronic wounds, and (2) to classify each clinical decision support model according to selected variables and to create an overview.

Methods

A systematic review was conducted using 6 databases. This systematic literature review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement for systematic reviews. The search strategy consisted of three facets, respectively: Facet 1 (Algorithm), Facet 2 (Wound care) and Facet 3 (Clinical decision support system). Studies based on acute wounds or trauma were excluded. Similarly, studies that presented guidelines, protocols and instructions were excluded, since they do not require progression along an active chain of reasoning from the clinicians, just their focus. Finally, studies were excluded if they had not undergone a peer review process. The following aspects were extracted from each article: authors, year, country, the sample size of data and variables describing the type of clinical decision support models. The decision support models were classified in 2 ways: quantitative decision support models, and qualitative decision support models.

Results

The final number of studies included in the systematic literature review was 10. These clinical decision support models included 4/10 (40%) quantitative decision support models and 6/10 (60%) qualitative decision support models. The earliest article was published in 2007, and the most recent was from 2015.

Conclusions

The clinical decision support models were targeted at a variety of different types of chronic wounds. The degree of accessibility of the inference engines varied. Quantitative models served as the engine and were invisible to the health care professionals, while qualitative models required interaction with the user.

The Importance of Nonlinear Transformations Use in Medical Data Analysis

Background

The accumulation of data and its accessibility through easier-to-use platforms will allow data scientists and practitioners who are less sophisticated data analysts to get answers by using big data for many purposes in multiple ways. Data scientists working with medical data are aware of the importance of preprocessing, yet in many cases, the potential benefits of using nonlinear transformations is overlooked.

Objective

Our aim is to present a semi-automated approach of symmetry-aiming transformations tailored for medical data analysis and its advantages.

Methods

We describe 10 commonly encountered data types used in the medical field and the relevant transformations for each data type. Data from the Alzheimer’s Disease Neuroimaging Initiative study, Parkinson’s disease hospital cohort, and disease-simulating data were used to demonstrate the approach and its benefits.

Results

Symmetry-targeted monotone transformations were applied, and the advantages gained in variance, stability, linearity, and clustering are demonstrated. An open source application implementing the described methods was developed. Both linearity of relationships and increase of stability of variability improved after applying proper nonlinear transformation. Clustering simulated nonsymmetric data gave low agreement to the generating clusters (Rand value=0.681), while capturing the original structure after applying nonlinear transformation to symmetry (Rand value=0.986).

Conclusions

This work presents the use of nonlinear transformations for medical data and the importance of their semi-automated choice. Using the described approach, the data analyst increases the ability to create simpler, more robust and translational models, thereby facilitating the interpretation and implementation of the analysis by medical practitioners. Applying nonlinear transformations as part of the preprocessing is essential to the quality and interpretability of results.

Structured Reporting in Neuroradiology: Intracranial Tumors

Purpose

The aim of this pilot study was to assess the clinical feasibility, diagnostic yield, advantages, and disadvantages of structured reporting for routine MRI-reading in patients with primary diagnosis of intracranial tumors as compared to traditional neuroradiological free text reporting.

Methods

A structured MRI reporting template was developed covering pathological, anatomical, and functional aspects in an itemized fashion. Retrospectively, 60 consecutive patients with first diagnosis of an intracranial tumor were selected from the radiology information system/PACS system. Structured reporting was performed by a senior neuroradiologist, blinded to clinical and radiological data. Reporting times were measured per patient. The diagnostic content was compared to free text reporting which was independently performed on the same MRI exams by two other neuroradiologists. The comparisons were categorized per item as: "congruent," "partially congruent," "incongruent," or "not mentioned in free-style report."

Results

Tumor-related items: congruent findings were found for all items (17/17) with congruence rates ranging between 98 and 39% per item. Four items achieved congruence rates ≥90%, 5 items >80%, and 9 items ≥70%. Partially congruent findings were found for all items in up to 50% per item. Incongruent findings were present in 7/17 items in up to 5% per item. Free text reports did not mention 12 of 17 items (range 7-43% per item). Non-tumor-related items, including brain atrophy, microangiopathy, vascular pathologies, and various extracranial pathologies, which were not mentioned in free-text reports between 18 and 85% per item. Mean reporting time for structured reporting was 7:49 min (3:12-17:06 min).

Conclusion

First results showed that expert structured reporting ensured reliable detection of all relevant brain pathologies along with reproducible documentation of all predefined diagnostic items, which was not always the case for free text reporting. A mean reporting time of 8 min per patient seems clinically feasible.

[Smart eye data : Development of a foundation for medical research using Smart Data applications]

BACKGROUND

Smart Data means intelligent data accumulation and the evaluation of large data sets. This is particularly important in ophthalmology as more and more data are being created. Increasing knowledge and personalized therapies are expected by combining clinical data from electronic health records (EHR) with measurement data.

OBJECTIVE

In this study we investigated the possibilities to consolidate data from measurement devices and clinical data in a data warehouse (DW).

MATERIAL AND METHODS

An EHR was adjusted to the needs of ophthalmology and the contents of referral letters were extracted. The data were imported into a DW overnight. Measuring devices were connected to the EHR by an HL7 standard interface and the use of a picture archiving and communications system (PACS). Data were exported from the review software using a self-developed software. For data analysis the software was modified to the specific requirements of ophthalmology.

RESULTS

In the EHR 12 graphical user interfaces were created and the data from 32,234 referral letters were extracted. A total of 23 diagnostic devices could be linked to the PACS and 85,114 optical coherence tomography (OCT) scans, 19,098 measurements from IOLMaster as well as 5,425 pentacam examinations were imported into the DW including over 300,000 patients. Data discovery software was modified providing filtering methods.

CONCLUSION

By building a DW a foundation for clinical and epidemiological studies could be implemented. In the future, decision support systems and strategies for personalized therapies can be based on such a database.

Outcomes after bariatric surgery according to large databases: a systematic review

PURPOSE

The rapid development of technological tools to record data allows storage of enormous datasets, often termed "big data". In the USA, three large databases have been developed to store data regarding surgical outcomes: the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP), the Healthcare Cost and Utilization Project (HCUP) National Inpatient Sample (NIS) and the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP). We aimed to evaluate the clinical impact of studies found in these databases concerning outcomes of bariatric surgery.

METHODS

We performed a systematic review using the Meta-analysis of Observational Studies in Epidemiology guidelines. Research carried out using the PubMed database identified 362 papers. All outcomes related to bariatric surgery were analysed.

RESULTS

Fifty-four studies, published between 2005 and February 2017, were included. These articles were divided into (1) outcomes related to surgical techniques (12 articles), (2) morbidity and mortality (12), (3) 30-day hospital readmission (10), (4) outcomes related to specific diseases (11), (5) training (2) and (6) socio-economic and ethnic observations in bariatric surgery (7). Forty-two papers were based on data from ACS-NSQIP, nine on data from NIS and three on data from MBSAQIP.

CONCLUSIONS

This review provides an overview of surgical management and outcomes of bariatric surgery in the USA. Large databases offer useful complementary information that could be considered external validation when strong evidence-based medicine data are lacking. They also allow us to evaluate infrequent situations for which randomized control trials are not feasible and add specific information that can complement the quality of surgical knowledge.

Big Data in traumatic brain injury; promise and challenges

Traumatic brain injury (TBI) is a spectrum disease of overwhelming complexity, the research of which generates enormous amounts of structured, semi-structured and unstructured data. This resulting big data has tremendous potential to be mined for valuable information regarding the "most complex disease of the most complex organ". Big data analyses require specialized big data analytics applications, machine learning and artificial intelligence platforms to reveal associations, trends, correlations and patterns not otherwise realized by current analytical approaches. The intersection of potential data sources between experimental TBI and clinical TBI research presents inherent challenges for setting parameters for the generation of common data elements and to mine existing legacy data that would allow highly translatable big data analyses. In order to successfully utilize big data analyses in TBI, we must be willing to accept the messiness of data, collect and store all data and give up causation for correlation. In this context, coupling the big data approach to established clinical and pre-clinical data sources will transform current practices for triage, diagnosis, treatment and prognosis into highly integrated evidence-based patient care.

Medical big data: promise and challenges

The concept of big data, commonly characterized by volume, variety, velocity, and veracity, goes far beyond the data type and includes the aspects of data analysis, such as hypothesis-generating, rather than hypothesis-testing. Big data focuses on temporal stability of the association, rather than on causal relationship and underlying probability distribution assumptions are frequently not required. Medical big data as material to be analyzed has various features that are not only distinct from big data of other disciplines, but also distinct from traditional clinical epidemiology. Big data technology has many areas of application in healthcare, such as predictive modeling and clinical decision support, disease or safety surveillance, public health, and research. Big data analytics frequently exploits analytic methods developed in data mining, including classification, clustering, and regression. Medical big data analyses are complicated by many technical issues, such as missing values, curse of dimensionality, and bias control, and share the inherent limitations of observation study, namely the inability to test causality resulting from residual confounding and reverse causation. Recently, propensity score analysis and instrumental variable analysis have been introduced to overcome these limitations, and they have accomplished a great deal. Many challenges, such as the absence of evidence of practical benefits of big data, methodological issues including legal and ethical issues, and clinical integration and utility issues, must be overcome to realize the promise of medical big data as the fuel of a continuous learning healthcare system that will improve patient outcome and reduce waste in areas including nephrology.

Efficiency Analysis of an Interoperable Healthcare Operations Platform

(1) Develop an enterprise platform to unify isolated information, software applications and team members. (2) Assess the efficiency of one benefit of the platform through comparative testing of employee document retrieval times. (3) Evaluate the level of satisfaction among our target audience. We developed an infrastructure to integrate information throughout our practice and make it available on a unified, secure, and remotely accessible platform. We solicited our practice for volunteers to test the new system. All interested volunteers participated. Thirteen employees searched for the same four items in both the new system and our legacy systems. Testing was performed in the pre-deployment stage. In our evaluation, we introduced an innovative method to precisely and objectively obtain data through the use of a widely available tool which could be leveraged for a variety of other studies. On average, it took our participants 7 min and 48 s to find four assigned items in our legacy systems. It only took our volunteers 1 min and 1 s to find the same items with the new platform (p-value 0.002). On a scale of 10 being the highest level of satisfaction, participants ranked the new system to be 8.7 while the traditional system was ranked at 6.3. An overarching enterprise platform is critical due to the ability to unify otherwise isolated applications, people and documents. Because navigating a new system would be expected to take longer than a familiar one, we were surprised by the dramatically improved efficiency and satisfaction of our new interoperable platform compared to the status quo. Since this platform was evaluated in the pre-deployment stage, we expect results to improve with employee experience as well as ongoing enhancements.

European healthcare systems readiness to shift from ‘one-size fits all’ to personalized medicine

Personalized medicine (PM) is no longer an abstract healthcare approach. It has become a reality over the last years and is already successfully applied in the various medical fields. Although there are success stories of implementing PM, there are still many more opportunities to further implement and make full use of the potential of PM. We assessed the system readiness of healthcare systems in Europe to shift from the predominant ‘one size fits all’ healthcare approach to PM. We conclude that European healthcare systems are only partially ready for PM. Key challenges such as integration of big data, health literacy, reimbursement and regulatory issues need to be overcome in order to strengthen the implementation and uptake of PM.

Teaching Shared Decision Making to Family Medicine Residents: A Descriptive Study of a Web-Based Tutorial

BACKGROUND

DECISION+2, a Web-based tutorial, was designed to train family physicians in shared decision making (SDM) regarding the use of antibiotics for acute respiratory infections (ARIs). It is currently mandatory for second-year family medicine residents at Université Laval, Quebec, Canada. However, little is known about how such tutorials are used, their effect on knowledge scores, or how best to assess resident participation.

OBJECTIVE

The objective of our study was to describe the usage of this Web-based training platform by family medicine residents over time, evaluate its effect on their knowledge scores, and identify what kinds of data are needed for a more comprehensive analysis of usage and knowledge acquisition.

METHODS

We identified, collected, and analyzed all available data about participation in and current usage of the tutorial and its before-and-after 10-item knowledge test. Residents were separated into 3 log-in periods (2012-2013, 2013-2014, and 2014-2015) depending on the day of their first connection. We compared residents' participation rates between entry periods (Cochran-Armitage test), assessed the mean rank of the difference in total scores and category scores between pre- and posttest (Wilcoxon signed-rank test), and compared frequencies of each. Subsequent to analyses, we identified types of data that would have provided a more complete picture of the usage of the program and its effect on knowledge scores.

RESULTS

The tutorial addresses 3 knowledge categories: diagnosing ARIs, treating ARIs, and SDM regarding the use of antibiotics for treating ARIs. From July 2012 to July 2015, all 387 second-year family medicine residents were eligible to take the Web-based tutorial. Out of the 387 eligible residents, 247 (63.8%) logged in at least once. Their participation rates varied between entry periods, most significantly between the 2012-2013 and 2013-2014 cohorts (P=.006). For the 109 out of 387 (28.2%) residents who completed the tutorial and both tests, total and category scores significantly improved between pre- and posttest (all P values <.001). However, the frequencies of those answering correctly on 2 of the 3 SDM questions did not increase significantly (P>.99, P=.25). Distribution of pre- or posttest total and category scores did not increase between entry periods (all P values >.1). Available data were inadequate for evaluating the associations between the tutorial and its impact on the residents' scores and therefore could tell us little about its effect on increasing their knowledge.

CONCLUSION

Residents' use of this Web-based tutorial appeared to increase between entry periods following the changes to the SDM program, and the tutorial seemed less effective for increasing SDM knowledge scores than for diagnosis or treatment scores. However, our results also highlight the need to improve data availability before participation in Web-based SDM tutorials can be properly evaluated or knowledge scores improved.

Big data science: A literature review of nursing research exemplars

BACKGROUND

Big data and cutting-edge analytic methods in nursing research challenge nurse scientists to extend the data sources and analytic methods used for discovering and translating knowledge.

PURPOSE

The purpose of this study was to identify, analyze, and synthesize exemplars of big data nursing research applied to practice and disseminated in key nursing informatics, general biomedical informatics, and nursing research journals.

METHODS

A literature review of studies published between 2009 and 2015. There were 650 journal articles identified in 17 key nursing informatics, general biomedical informatics, and nursing research journals in the Web of Science database. After screening for inclusion and exclusion criteria, 17 studies published in 18 articles were identified as big data nursing research applied to practice.

DISCUSSION

Nurses clearly are beginning to conduct big data research applied to practice. These studies represent multiple data sources and settings. Although numerous analytic methods were used, the fundamental issue remains to define the types of analyses consistent with big data analytic methods.

CONCLUSION

There are needs to increase the visibility of big data and data science research conducted by nurse scientists, further examine the use of state of the science in data analytics, and continue to expand the availability and use of a variety of scientific, governmental, and industry data resources. A major implication of this literature review is whether nursing faculty and preparation of future scientists (PhD programs) are prepared for big data and data science.

The future of psychiatric research

Abstract Psychiatric disorders place considerable burden on individuals and on public health. Funding for research in psychiatry is less than ideal, but even so high quality research is being conducted at many centers. However, these studies have not impacted clinical practice as much as expected. The complexity of psychiatric disorders is one of the reasons why we face difficulties in translating research results to patient care. New technologies and improved methodologies are now available and must be incorporated to deal with this complexity and to accelerate the translational process. I discuss the application of modern techniques for data acquisition and analysis and also the new possibilities for performing trials in virtual models of biological systems. Adoption of new technologies is necessary, but will not reduce the importance of some of the fundamentals of all psychiatry research, such as the developmental and translational perspectives. Psychiatrists wishing to integrate these novelties into their research will need to work with contributors with whom they are unaccustomed to working, such as computer experts, a multidisciplinary team, and stakeholders such as patients and caregivers. This process will allow us to further understand and alleviate the suffering and impairment of people with psychiatric disorders.

Big Data in Health: a Literature Review from the Year 2005

The information stored in healthcare systems has increased over the last ten years, leading it to be considered Big Data. There is a wealth of health information ready to be analysed. However, the sheer volume raises a challenge for traditional methods. The aim of this article is to conduct a cutting-edge study on Big Data in healthcare from 2005 to the present. This literature review will help researchers to know how Big Data has developed in the health industry and open up new avenues for research. Information searches have been made on various scientific databases such as Pubmed, Science Direct, Scopus and Web of Science for Big Data in healthcare. The search criteria were "Big Data" and "health" with a date range from 2005 to the present. A total of 9724 articles were found on the databases. 9515 articles were discarded as duplicates or for not having a title of interest to the study. 209 articles were read, with the resulting decision that 46 were useful for this study. 52.6 % of the articles used were found in Science Direct, 23.7 % in Pubmed, 22.1 % through Scopus and the remaining 2.6 % through the Web of Science. Big Data has undergone extremely high growth since 2011 and its use is becoming compulsory in developed nations and in an increasing number of developing nations. Big Data is a step forward and a cost reducer for public and private healthcare.

Is real world evidence influencing practice? A systematic review of CPRD research in NICE guidances

BACKGROUND

There is currently limited evidence regarding the extent Real World Evidence (RWE) has directly impacted the health and social care systems. The aim of this review is to identify national guidelines or guidances published in England from 2000 onwards which have referenced studies using the governmental primary care data provider the Clinical Practice Research Datalink (CPRD).

METHODS

The methodology recommended by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was followed. Four databases were searched and documents of interest were identified through a search algorithm containing keywords relevant to CPRD. A search diary was maintained with the inclusion/exclusion decisions which were performed by two independent reviewers.

RESULTS

Twenty-five guidance documents were included in the final review (following screening and assessment for eligibility), referencing 43 different CPRD/GPRD studies, all published since 2007. The documents covered 12 disease areas, with the majority (N =7) relevant to diseases of the Central Nervous system (CNS). The 43 studies provided evidence of disease epidemiology, incidence/prevalence, pharmacoepidemiology, pharmacovigilance and health utilisation.

CONCLUSIONS

A slow uptake of RWE in clinical and therapeutic guidelines (as provided by UK governmental structures) was noticed. However, there seems to be an increasing trend in the use of healthcare system data to inform clinical practice, especially as the real world validity of clinical trials is being questioned. In order to accommodate this increasing demand and meet the paradigm shift expected, organisations need to work together to enable or improve data access, undertake translational and relevant research and establish sources of reliable evidence.

Computational Electrocardiography: Revisiting Holter ECG Monitoring

BACKGROUND

Since 1942, when Goldberger introduced the 12-lead electrocardiography (ECG), this diagnostic method has not been changed.

OBJECTIVES

After 70 years of technologic developments, we revisit Holter ECG from recording to understanding.

METHODS

A fundamental change is fore-seen towards "computational ECG" (CECG), where continuous monitoring is producing big data volumes that are impossible to be inspected conventionally but require efficient computational methods. We draw parallels between CECG and computational biology, in particular with respect to computed tomography, computed radiology, and computed photography. From that, we identify technology and methodology needed for CECG.

RESULTS

Real-time transfer of raw data into meaningful parameters that are tracked over time will allow prediction of serious events, such as sudden cardiac death. Evolved from Holter's technology, portable smartphones with Bluetooth-connected textile-embedded sensors will capture noisy raw data (recording), process meaningful parameters over time (analysis), and transfer them to cloud services for sharing (handling), predicting serious events, and alarming (understanding). To make this happen, the following fields need more research: i) signal processing, ii) cycle decomposition; iii) cycle normalization, iv) cycle modeling, v) clinical parameter computation, vi) physiological modeling, and vii) event prediction.

CONCLUSIONS

We shall start immediately developing methodology for CECG analysis and understanding.

Big Data in medical research and EU data protection law: challenges to the consent or anonymise approach

Medical research is increasingly becoming data-intensive; sensitive data are being re-used, linked and analysed on an unprecedented scale. The current EU data protection law reform has led to an intense debate about its potential effect on this processing of data in medical research. To contribute to this evolving debate, this paper reviews how the dominant 'consent or anonymise approach' is challenged in a data-intensive medical research context, and discusses possible ways forwards within the EU legal framework on data protection. A large part of the debate in literature focuses on the acceptability of adapting consent or anonymisation mechanisms to overcome the challenges within these approaches. We however believe that the search for ways forward within the consent or anonymise paradigm will become increasingly difficult. Therefore, we underline the necessity of an appropriate research exemption from consent for the use of sensitive personal data in medical research to take account of all legitimate interests. The appropriate conditions of such a research exemption are however subject to debate, and we expect that there will be minimal harmonisation of these conditions in the forthcoming EU Data Protection Regulation. Further deliberation is required to determine when a shift away from consent as a legal basis is necessary and proportional in a data-intensive medical research context, and what safeguards should be put in place when such a research exemption from consent is provided.

Creation of an Accurate Algorithm to Detect Snellen Best Documented Visual Acuity from Ophthalmology Electronic Health Record Notes

Background

Visual acuity is the primary measure used in ophthalmology to determine how well a patient can see. Visual acuity for a single eye may be recorded in multiple ways for a single patient visit (eg, Snellen vs. Jäger units vs. font print size), and be recorded for either distance or near vision. Capturing the best documented visual acuity (BDVA) of each eye in an individual patient visit is an important step for making electronic ophthalmology clinical notes useful in research.

Objective

Currently, there is limited methodology for capturing BDVA in an efficient and accurate manner from electronic health record (EHR) notes. We developed an algorithm to detect BDVA for right and left eyes from defined fields within electronic ophthalmology clinical notes.

Methods

We designed an algorithm to detect the BDVA from defined fields within 295,218 ophthalmology clinical notes with visual acuity data present. About 5668 unique responses were identified and an algorithm was developed to map all of the unique responses to a structured list of Snellen visual acuities.

Results

Visual acuity was captured from a total of 295,218 ophthalmology clinical notes during the study dates. The algorithm identified all visual acuities in the defined visual acuity section for each eye and returned a single BDVA for each eye. A clinician chart review of 100 random patient notes showed a 99% accuracy detecting BDVA from these records and 1% observed error.

Conclusions

Our algorithm successfully captures best documented Snellen distance visual acuity from ophthalmology clinical notes and transforms a variety of inputs into a structured Snellen equivalent list. Our work, to the best of our knowledge, represents the first attempt at capturing visual acuity accurately from large numbers of electronic ophthalmology notes. Use of this algorithm can benefit research groups interested in assessing visual acuity for patient centered outcome. All codes used for this study are currently available, and will be made available online at https://phekb.org.

Utilizing electronic health records to predict acute kidney injury risk and outcomes: workgroup statements from the 15(th) ADQI Consensus Conference

The data contained within the electronic health record (EHR) is "big" from the standpoint of volume, velocity, and variety. These circumstances and the pervasive trend towards EHR adoption have sparked interest in applying big data predictive analytic techniques to EHR data. Acute kidney injury (AKI) is a condition well suited to prediction and risk forecasting; not only does the consensus definition for AKI allow temporal anchoring of events, but no treatments exist once AKI develops, underscoring the importance of early identification and prevention. The Acute Dialysis Quality Initiative (ADQI) convened a group of key opinion leaders and stakeholders to consider how best to approach AKI research and care in the "Big Data" era. This manuscript addresses the core elements of AKI risk prediction and outlines potential pathways and processes. We describe AKI prediction targets, feature selection, model development, and data display.