PSI One Day Meeting: Beyond the detail seeing the whole picture; Graphical representations, what’s in it for us?
Data visualisation is one of the fundamental tools that we have to explore, understand and present data. However, do we really put enough effort and time into this aspect of our work? This PSI one day meeting aims to cover a wide variety of areas where graphics can, and are, being used within the pharmaceutical industry and beyond. By sharing and engaging in multiple and varied examples attendees will come away from the meeting with a renewed appreciation of the benefits of graphical representations and ideas of where these tools could prove useful within their own areas of work.
Name of talks
Slides to download
The Clinical Data Cinema: The many roles of data visualisation in clinical developmentJürgen Löffler, (Novartis)
Please click here to see the Individual Patient Example
The Clinical Data Cinema: The many roles of data visualisation in clinical development
Abstract: The role of data visualization in various aspects of clinical data analysis and reporting will be illustrated. Examples are Clinical Study Reports and submission dossiers, benefit-risk evaluations, slide presentations, data exploration, statistical model diagnostics, and ongoing review of collected study data. Possibilities of how to present data, along with specific requirements as well as limitations and constraints in each area will be discussed. Finally, a structured approach towards definition of the appropriate graph for certain purposes and data types will be presented.
Juergen Loeffler (Novartis)
Riding the Data Explosion
Abstract: Visualisation of statistical analysis has been a key part of clinical trials for many years. Now it is becoming possible to visually explore the statistical analysis of colossal datasets in real-time using multi-machine in-memory engines. This is a seismic shift in the ability to explore and discover new information - and more importantly - by potentially different people. There is an opportunity for those with statistical analysis skills to be at the vanguard of this explosion in understanding and to shape the way in which new indications are found and the value of the assets of the pharmaceutical companies are maximised.
Simon Tilley (SAS)
Visualisation of Clinical Trial Data Using Animated Graphics
Abstract: Conventional data visualisation uses static images. An innovative visualisation method using animated graphics could help to review, analyse and communicate large amounts of information effectively. In 1930, astronomer Clyde Tombaugh discovered Pluto byswitching back and forth between images of the night sky to look for moving objects. In 2010, Prof Hans Rosling used animated graphics to teach the world about health and economics. This talk will discuss the potential application of this approach to medical data analysis, and introduce a new software tool that turns clinical trial data into animated graphics.
Jianmei Wang (Roche Products Limited)
The Rise and Rise of Graphic Expectations
Abstract: It was once the case that a simple table was sufficient for displaying information but the expectations for graphics we see and produce has grown rapidly. We are all used to seeing highly customised interactive graphics on a daily basis but how can we use these tools effectively for our own exploratory analysis and reporting? In this talk we will look at how expectations have increased and how we can effectively use tools available to us for presenting information.
Aimee Gott (Mango Solutions)
Interactive Statistical Graphics: When charts come to life
Abstract: There is a comprehensive suite of univariate and multivariate statistical graphics, covering data on categorical, continuous and mixed scales. Whereas many of these graphics are widely known and used for diagnostics and presentation of models and raw data (cf. the popular R package ggplot2), adding interactive features may make these graphics to versatile tools in exploratory data analysis.
The most common interactive feature is the selection of subsets of data to focus on "what's interesting". Linked highlighting throughout all linked graphs of the same data can highlight these features in even more dimensions. Linked-highlighting can efficiently be used to compare subgroups within a dataset, or to investigate the conditional distribution of subgroups. Further interactions include the instant change of plot parameters and plot scales, as well as the flexible incorporation of statistical information into standard statistical graphics.
This talk illustrates the various interaction techniques and how these techniques facilitate the analysis of data with purely graphical means. Most of these techniques are demonstrated within the graphical data analysis software Mondrian (http://www.theusRus.de/Mondrian)
Martin Theus (Telefonica Germany)
Every Picture Tells a Story: Using many graphics at once
Abstract: People are very skilled at scanning large amounts of information and picking out the bits that are important to them—when they are experienced in doing this. It is now easy to draw many informative graphics quickly and flexibly, but typically people don’t. They tend to draw individual graphics and use them more for presenting information than for gaining information. This talk discusses using collections of graphics to find information and to generate ideas to be modelled and tested. Several applied examples will be presented and audience participation in evaluating the graphics and the graphical analyses will be welcome.
Antony Unwin (University of Ausburg)
Using Graphics to Present the Study Results as a Story
Abstract: Too often we watch presentations of the final results of a study that has just been reported and think of how it can be more interesting. We are told that we need to see the numbers, the tables, the figures, so that we know the result of the study. However, when we sit and listen to the results of a study, what we are really after is a story. We want the data to be converted to knowledge, something we can walk away with. We don’t walk away with tables, and numbers, but we do remember a story told by a good story teller, with all the pictures to visualise what had happened in the study. This presentation will show how we can use graphics to tell a story, converting Data to Knowledge the audience of any presentation can walk away with.
Shafi Chowdhury (Shafi Consultancy)
Graphical Analysis for Epidemiology, Manufacturing and Marketing Applications
Abstract: Much attention is given to graphics for analysis and reporting in R&D. In this paper, I provide a review of graphics used in the other functional areas including sales, marketing and manufacturing. Examples are drawn from the life sciences, retail and consumer packaged goods industries. Graphical methods for exploratory review and publication reporting are referenced.
Michael O’Connell (Tibco), Andrew Berridge (Tibco)
Joint PSI, EFSPI & ASA BIOP Webinar: Estimands
PSI, the European Federation of Statisticians in the Pharmaceutical Industry (EFSPI) and the Biopharmaceutical Section of the American Statistical Association (ASA) are jointly organising a webinar on Estimands in Practice. Speakers from regulatory authorities (FDA and EMA) and industry will present on their experience on this topic to date.
Discover your potential by working as a Covance Senior Principal / Principal Biostatistician. You'll enjoy a varied role working for different sponsors across several therapeutic areas, leading projects and mentoring junior members of the department.
Biometrics at AstraZeneca provides the data that influences decisions on how we roll back the frontiers of science to bring life-changing medicines to the world. We are at the heart of design, analytics and interpretation of AstraZeneca’s portfolio.