Working Group Overview
New challenges in regulatory science and drug, biologic, and device development provide new opportunities for recognizing and leveraging new or emerging technologies and computational tools or underutilized existing technologies. Initiated at the 2013 PhUSE Annual CSS, the Emerging Technologies working group provides a forum for determining interest in specific computational science topics, tools, technologies, and approaches.
This Emerging Technologies working group will be an open, transparent forum for sharing pre-competitive means of applying new technologies and is being challenged with creation of well-defined collaborative projects that will describe, prioritize, assess, and assist advancement of these opportunities. Possible topics include (but are not limited to) semantic web applications, analysis metadata, modeling, simulation, and “The Cloud”. Projects incorporating these topics might include prioritization, development, and piloting for feasibility and value.
Current Leadership Team
|Andy Richardson||ET Co-lead||d-Wisefirstname.lastname@example.org|
|Sam Hume||CDICS Liaison||CDISCemail@example.com|
|Project||Description||Recruiting New Members|
|Cloud Adoption in the Life Sciences Industry||Cloud technology and its use of multi-tennant app solutions are increasing the capabilities of Life Sciences solutions and reducing IT infrastructure costs through the sharing of infrastructure and investment cross-industry. The goal of this work stream has been to provide a practical, usable framework to overcome those barriers. Through the use of this framework, it is envisaged that the barriers to adoption by pharma of cloud-based technology will be addressed.||Yes|
|Data Visualisations for Clinical Data||The FDA Guidance on a Risk-Based Approach to Monitoring (August 2013) opened the door to using scientifically founded monitoring solutions as alternatives to 100% source verification of clinical data. Individual companies have proposed a range of opportunities to look at the applicability of data visualization within the pharmaceutical environment that addresses cross-domain questions and insight associated with RBM.||Yes|
|Investigating the use of FHIR in Clinical Research||Increasing interest in eSource keeps the issue of data integration between Research Systems (EDC, CTMS, CDMS, etc) and healthcare systems (EHR, etc) as a consistent want for Sponsors clinical investigators and Regulators. The new PhUSE project 'Evaluating the Use of FHIR in Clinical Research' will look at how the HL7 FHIR standard could be used as a fundamental part of the clinical trial process in the future.||Yes|
|BlockChain Technology||Introduce Blockchain and describe how it works. Pre-requisites to adopt Blockchain. Understand the qualities of Blockchain relevant to the Pharma setting and the example of use cases and applications. Provide high level analysis of Pros/Cons of BlockChain in Pharma and Healthcare.||Yes|
|ODM4 Submissions||The Operational Data Model (ODM) standard has been the CDISC standard format for data exchange since 2000. Define-XML and Dataset-XML are ODM extensions supporting the transport of CDISC dataset metadata and data, respectively. Define-XML is now a required part of a regulatory submission. However, despite using Define-XML to submit dataset metadata, other data and metadata required for submissions are submitted in different file formats that adversely restrict data representation, machine-readability options, and the ability to validate submissions.||Yes|
|Going Translational with Linked Data||This project builds on the successful completion of the "Clinical Trials Data as RDF" project where four SDTM domains (DM, VS, EX, TS) were modeled in RDF, and the ontologies used to create RDF instance data.Existing domains will be broadened to include non-clinical concepts, thus extending the impact of the project further along the data lifecycle. A minimum of two additional domains will be added, starting with AE (and non-clinical AE equivalent observations).||Yes|
|Key Performance Indicators & Metrics||Collecting, tracking and evaluating data on an ongoing basis can provide organizations with credible, compelling information when communicating with key decision-makers and stake-holders.The PhUSE Data Science & KPI Metric Reporting Group has been working to establish a set of common Data Reporting Metrics which are more detailed than industry wide metrics, therefore, allowing a greater level of granularity in our project reporting, and business process management|| Yes
|Machine Learning/Artificial Intelligence||The most popular buzz word nowadays in the technology world is 'Machine Learning' (ML) and Artificial Intelligence (AI). Most economists and business experts foresee ML & AI changing every aspect of our lives in the next 10 years through automating and optimizing processes such as self-driving vehicles, online recommendations on Netflix and Amazon, fraud detection in banks, image and video recognition, natural language processing, question answering machines (e.g., IBM Watson) and many more.||Yes|
- Metadata Definition Project: File:ET-project-template-metadata-definitions.doc
- Metadata Management Project: File:ET-project-Template - User Guide for E2E meta data management.doc
After the Emerging Technology Round Table Sessions the following projects were identified as prospective working groups for the 2015 CSS and beyond. Some of these need passionate people to lead!
|Project||Description||Sponsor||Needs a Lead|
|Framework for mHealth||As personal fitness trackers become more ubiquitous, industry is starting to assess how these devices can be used to supplement the clinical record. Very much in its infancy, use of these personal devices is a grey area in terms of adoption status, information quality (for example, the use of proprietary data gathering algorithms), patient privacy and informed consent. This group will develop a framework so companies looking to adopt these devices in their clinical trials can make an informed decision as they assess the potential of integrating mobile health with clinical research.||Tony Hewer||Maybe|
|Machine Learning for Clinical Research||With the continually raising profile of Data Science in the Pharmaceutical Industry, techniques which have traditionally not been widely adopted in the industry are starting to become part of a clinical analyst's toolbox. Text Analytics, already employed in other industries, enables analysts to uncover clinically important information in previously opaque free-text. Techniques to be discussed include sentiment analysis, concept/entity extraction, clustering and other approaches. This group will gather people interested in text mining techniques to develop frameworks and approaches for extracting structured information from unstructured clinical trial text data.||CDER||Yes|
|Big Data approaches in Clinical Research||Big Data is a term we are all familiar with - it is promoted as the way to comprehensively answer multiple questions across multiple domains. But what is Big Data? Is it relevant to clinical researcher? Is it simply large volumes of data? Is it these large datasets plus the techniques needed to extract value from this data? What approaches work and when? This group will begin by answering the most basic questions associated with Big Data and clinical research -- is it relevant, and where do we go from here? This topic has been previously posed and we have a Project Request: File:ET-project-Template-BigData.doc||N/A||Yes|
Projects on Hold
|Project||Description||ODM4 Submissions||The Operational Data Model (ODM) standard has been the CDISC standard format for data exchange since 2000. Define-XML and Dataset-XML are ODM extensions supporting the transport of CDISC dataset metadata and data, respectively. Define-XML is now a required part of a regulatory submission. However, despite using Define-XML to submit dataset metadata, other data and metadata required for submissions are submitted in different file formats that adversely restrict data representation, machine-readability options, and the ability to validate submissions.|