Video-on-Demand Library


Collection
Keyword or phrase

Topic

08 December 2021

Demographic data display is a highly recurrent issue. Rhys Warham presents proposals how a comprehensive graphical display of demographic data could look like. Visualisations are available on the Wonderful Wednesday blog.

Read more...

Demographic data display is a highly recurrent issue. Rhys Warham presents proposals how a comprehensive graphical display of demographic data could look like. Visualisations are available on the Wonderful Wednesday blog

Demographic data often consists of multiple characteristics with different data types. How to effectively combine these? One way to even include individual values is using beeswarm plots. Using scatterplots also gives the possibility to use color to display categorical data. It is advisable to use colorblind proof  palette. A universal tool for a quick and easy look at the data is Describer. Another tool presents plots and explaining text side by side. And for a comprehensive overview facet boxplots are a tool that works without a lot of explanation. Enjoy looking!

There’s no challenge for January. Instead the first webinar in 2022 will give a step by step guidance how to improve a given graph making use of gestalt principles. See the Wonderful Wednesday homepage for more detail.

Wonderful Wednesdays are brought to you by the Visualisation SIG. The Wonderful Wednesday team includes: Bodo Kirsch, Alexander Schacht, Mark Baillie, Daniel Saure, Zachary Skrivanek, Lorenz Uhlmann, Rachel Phillips, Markus Vogler, David Carr, Steve Mallett, Abi Williams, Julia Igel, Gakava Lovemore, Katie Murphy, Rhys Warham, Sara Zari, Irene de la Torre Arenas

02 December 2021

The first talk is about "Introduction to Statistical Methods and Challenges in Vaccine Development" by Ivan Chan and the second talk about "Vaccine quintet. Statistical issues in the design and analysis of five vaccine programmes" by Stephen Senn.

Read more...

The first talk is about "Introduction to Statistical Methods and Challenges in Vaccine Development" by Ivan Chan and the second talk about "Vaccine quintet. Statistical issues in the design and analysis of five vaccine programmes" by Stephen Senn.

Introduction to Statistical Methods and Challenges in Vaccine Development.
Vaccines have long been recognized as one of the greatest achievements in public health. Through mass immunization, smallpox was eradicated globally in 1979 and polio was eliminated in the Americas in 1994. In the fight against the COVID-19 pandemic, several vaccines have been successfully developed in record time thanks to incredible scientific innovation and public-private partnership. A new vaccine must pass high bars of safety and efficacy to justify the risk benefit of potential mass immunization. In this presentation, we will first introduce the concept of immunity and explain how vaccine works in stimulating immune responses and protecting people from diseases. Then we will give an overview of the statistical methods commonly used in assessing vaccine’s efficacy and safety. In addition, we will discuss a few special challenges in developing vaccines and give examples of statistical innovation that have helped accelerate the development of novel vaccines.

Vaccine quintet. Statistical issues in the design and analysis of five vaccine programmes.
The response to the COVID-19 crisis by various vaccine developers has been extraordinary, both in terms of speed of response and the delivered efficacy of the vaccines. It has also raised some fascinating issues of design, analysis and interpretation. I shall consider some of these issues, taking as my example, five vaccine programmes: Pfizer/BioNTech, AstraZeneca/Oxford, Moderna, Novavax, and J&J Janssen paying particular attention to the first two. Among matters covered will be concurrent control, efficient design, issues of measurement raised by two-shot vaccines and implications for roll-out, and the surprising effectiveness of simple analyses. Differences between the five development programmes as they affect statistics will be covered but some essential similarities will also be discussed. A key issue is the difference between causal and predictive inference, a matter that has become important due to the emergence of new viral variants on the one hand and (of necessity) limited follow-up in the clinical trials that were conducted.

19 November 2021

Building on the previous webinar series, we will present a refresher on the key elements of the estimand framework and use the ETHOS trial (NCT02465567) to illustrate the estimand framework and to highlight the different options that may be of interest to different stakeholders. The background and disease context of the case will be described, the important events (termed intercurrent events) occurring in the trial will be described and three estimands will be showcased. It will then be explored how best to communicate the estimands in medical journals. Topics of discussion will be raised at different points during the training and there will be time allowed (30 mins) to answer your questions.

Read more...

Building on the previous webinar series, we will present a refresher on the key elements of the estimand framework and use the ETHOS trial (NCT02465567) to illustrate the estimand framework and to highlight the different options that may be of interest to different stakeholders. The background and disease context of the case will be described, the important events (termed intercurrent events) occurring in the trial will be described and three estimands will be showcased. It will then be explored how best to communicate the estimands in medical journals. Topics of discussion will be raised at different points during the training and there will be time allowed (30 mins) to answer your questions.

Contents of Webinar Topics:
06:40
Learning Outcomes
07:30 Introduction to the ETHOS study
16:14 Reminder of the Estimand Framework
21:56 Using the Estimand Framework with ETHOS
26:00 3 possible estimands
34:30 Clinical view
39:13 How should estimands be communicated in medical journals
48:07 Conclusions and Recap Learning Outcomes
51:10 General Q & A

 

17 November 2021

In this webinar we look at some recent advances in statistical methods for identifying treatment effect heterogeneity in clinical trials. This ranges from identifying baseline biomarkers likely to influence the treatment effect (ranking) to provide novel biomarker 'signatures' (subgroups) with associated estimated enhanced effect (Individual Treatment Effects).

Read more...

David Svensson (AstraZeneca), Ilya Lipkovich (Eli Lilly), Björn Bornkamp (Norvatis), Kostas Sechidis (Novartis), Paolo Eusebi (UCB Pharma)

In this webinar we look at some recent advances in statistical methods for identifying treatment effect heterogeneity in clinical trials. This ranges from identifying baseline biomarkers likely to influence the treatment effect (ranking) to provide novel biomarker 'signatures' (subgroups) with associated estimated enhanced effect (Individual Treatment Effects). Some practical issues ranges from overfitting risks, biases, and confounding of prognostic and predictive effects. Modern methods aim to overcome such potential difficulties while remaining flexible, and offer a structured approach to the problem (aiming to avoid the notorious 'data dredging'). The novel techniques are often tree based and/or penalized regression, i.e., with a machine learning flavour. Sometimes the aim of the analysis is to predict the individual optimal treatment allocation given baseline biomarker data (Individual Treatment Rules). Efficient Visualization of relationships in the data is also of importance in the practical applications. The talks will highlight and discuss such aspects and will also reflect typical aspects discussed within the EFSPI/PSI Subgroup Special Interest Group. (While this event is not intended as a formal course, it will still serve as an introduction and overview to the area, as well as covering some more technically challenging material for the more experienced participant).

10 November 2021

Understanding correlations is one of the most important things in data science. How can effective visualisations can be helpful? Bodo Kirsch leads the discussion on different displays to combine the clinical outcome with the anchor measurement. Visualisations are available on the Wonderful Wednesday blog.

Read more...

Understanding correlations is one of the most important things in data science. How can effective visualisations can be helpful? Bodo Kirsch leads the discussion on different displays to combine the clinical outcome with the anchor measurement. Visualisations are available on the Wonderful Wednesday blog.

The presented visualisations are all about explaining the link between a clinical symptom score and the anchor measurement CGI-I. The first approach is to show cumulative distribution functions and prediction intervals grouped by the CGI-I score. Inside into the individual data is given by a comprehensive display including paired distribution function plot with integrated line plots. But there’s much more to it – a must see. More examples show different ways to display distribution functions in a way that visually highlights important differences. And last but not least this is given as a story-telling example. The challenge for December is about displaying demographic data based on an example data set on Alzheimer patients. See the Wonderful Wednesday homepage for more detail.

Wonderful Wednesdays are brought to you by the Visualisation SIG. The Wonderful Wednesday team includes: Bodo Kirsch, Alexander Schacht, Mark Baillie, Daniel Saure, Zachary Skrivanek, Lorenz Uhlmann, Rachel Phillips, Markus Vogler, David Carr, Steve Mallett, Abi Williams, Julia Igel, Gakava Lovemore, Katie Murphy, Rhys Warham, Sara Zari, Irene de la Torre Arenas

28 October 2021

Here, we show how to estimate potentially time-varying placebo-controlled vaccine efficacy in this type of staggered vaccination of participants. In addition, we compare the performance of blinded and unblinded crossover designs in estimating long-term vaccine efficacy.

Read more...

Prof. Dan-Yu Lin 

Large-scale deployment of safe and durably effective vaccines can curtail the COVID-19 pandemic. However, the high vaccine efficacy reported by ongoing phase 3 placebo-controlled clinical trials is based on a median follow-up time of only about two months and thus does not pertain to long-term efficacy. To evaluate the duration of protection while allowing trial participants timely access to efficacious vaccine, investigators can sequentially cross participants over from the placebo arm to the vaccine arm. Here, we show how to estimate potentially time-varying placebo-controlled vaccine efficacy in this type of staggered vaccination of participants. In addition, we compare the performance of blinded and unblinded crossover designs in estimating long-term vaccine efficacy.  

21 October 2021

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 Complex Innovative Designs (CID) in practice. Speakers from regulatory authorities and industry will present on their experience.

Read more...

John Scott, Dieter Häring, Marius Thomas, Olivier Collignon

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 Complex Innovative Designs (CID) in practice. Speakers from regulatory authorities and industry will present on their experience.
The following aspects are covered:
• Overview of the FDA Complex Innovative Trial Design pilot program and the applications received to date together with details on some of them
• Overview of the FDA guidance on interacting on Complex Innovative Trial Designs
• Detailed case study of a clinical trial in children which was evaluated within FDA’s CID pilot program, applying borrowing of information from external trials in adults
• Overview of statistical and regulatory considerations on master protocols, focusing on Phase III confirmatory trials

21 October 2021

Imposter Syndrome is experienced by 70% of people and the feelings of self-doubt and inadequacy it promotes can hold us back from valuing our accomplishments and reaching our full potential. Learn from others in the PSI Community who also associate with Imposter Syndrome.

Read more...

Kimberley Hacquoil, Chrissie Fletcher, Paul Terrill, Claire Brittain, Anashua Banerji

Do you ever think: “I don’t belong here, I’m not as capable as they think I am” “I don’t deserve it, I’m just lucky” “I'm never good enough, others are better” “I’m a fraud, I’ll get found out”. Imposter Syndrome is experienced by 70% of people and the feelings of self-doubt and inadequacy it promotes can hold us back from valuing our accomplishments and reaching our full potential. Learn from others in the PSI Community who also associate with Imposter Syndrome. They will discuss the topic, give their own personal views and experiences and provide tips to beat it.

13 October 2021

Bodo Kirsch discusses an approach to display data on competing risks based on an issue recently published. Visualisations are available on the Wonderful Wednesday blog.

Read more...

Bodo Kirsch discusses an approach to display data on competing risks based on an issue recently published. Visualisations are available on the Wonderful Wednesday blog

The proposed visualization by Agustin Calatroni consists of two parts. The first one gives a general overview on baseline parameters in a graphic-table-combination  (grable) with interactive histograms. The second part shows the impact of this parameters on the different risks in another grable with expandable CIF display. Different possibilities for layout, coloring and description are discussed.

The challenge for November is about displaying clinically relevant differences based on a data example with two scores. How to visually communicate relevant changes in one score using the second as anchor? See the Wonderful Wednesday homepage for more detail. 

Wonderful Wednesdays are brought to you by the Visualisation SIG. The Wonderful Wednesday team includes: Bodo Kirsch, Alexander Schacht, Mark Baillie, Daniel Saure, Zachary Skrivanek, Lorenz Uhlmann, Rachel Phillips, Markus Vogler, David Carr, Steve Mallett, Abi Williams, Julia Igel, Gakava Lovemore, Katie Murphy, Rhys Warham, Sara Zari, Irene de la Torre Arenas

17 September 2021

Gain new insights into data visualisations and learn about the regulatory perspective with this event, which provides a practical introduction to modern methods of data visualisation, including presentations from some well-known and influential speakers.

Read more...

Data visualisation has been used to gain insights into medical data for over 150 years. More modern methods include interactive and animated data visualisation tools, and the development of open source code using agile methods. Gain new insights into data visualisations and learn about the regulatory perspective with this event, which provides a practical introduction to modern methods of data visualisation, including presentations from some well-known and influential speakers.

Key Timestamps:

00:16 Welcome & Introduction
04:00 Susan Mayo - Making impactful graphs: Looking through the eyes of your audience
58:00 Jeremy Wildfire - Building Open-Source Tools for Safety Monitoring 
1:39:40 Sheila Dickinson - Points to bear in mind for visual displays of benefit-risk data
2:07:50 Matthias Trampisch - Dynamic data visualization for Benefit/Risk Assessment DURING trial conduct
2:15:30 Charlotta Fruechtenicht - visR: A Package for Effective Visualizations in Pharma
2:33:35 Patrick Schlömer - A concept to foster interactive and interdisciplinary data investigations using intelligent visualizations
3:00:10 Panel Q&A Discussion

08 September 2021

The Wonderful Wednesday of September is about the display of treatment differences in time-series data on the example of Continuous Glucose Monitoring (CGM) data. Zachary Skrivanek presents the visualisations proposed for this challenger ranging from explanatory plots with condensed information to an exploratory interactive dashboard. All visualisations are available on the Wonderful Wednesday blog.

Read more...

The Wonderful Wednesday of September is about the display of treatment differences in time-series data on the example of Continuous Glucose Monitoring  (CGM) data. Zachary Skrivanek presents the visualisations proposed for this challenger ranging from explanatory plots with condensed information to an exploratory interactive dashboard. All visualisations are available on the Wonderful Wednesday blog.

The main message of successful glycemic control is nicely transported in a box plot using intuitive coloring. For a detailed look at the subject level data the CGM dashboard gives the possibility to combine different displays while being able to sort and filter the data. In addition it plots the results of recursive partitioning models that can be interactively applied to a huge set of parameters.

Last but not least a set of different multi-plots are discussed highlighting pros and cons of overlayed bands, averages, modelled averages, and scatter plots. This includes the usage of various visual elements as well as effective decluttering of a visualisation.

The October challenge is on a competing risk in a recent COVID-19 data example. How to find a way to display the impact of different risk factors on the risk of death considering the competing risk of recovery? See the Wonderful Wednesday homepage for more detail.

Wonderful Wednesdays are brought to you by the Visualisation SIG. The Wonderful Wednesday team includes: Bodo Kirsch, Alexander Schacht, Mark Baillie, Daniel Saure, Zachary Skrivanek, Lorenz Uhlmann, Rachel Phillips, Markus Vogler, David Carr, Steve Mallett, Abi Williams, Julia Igel, Gakava Lovemore, Katie Murphy, Rhys Warham, Sara Zari, Irene de la Torre Arenas



11 August 2021

For this edition of the Wonderful Wednesdays, the audience was asked to send examples of visualizations representing coronavirus data. Mark Baillie guided the webinar, showcasing different types of visualizations, such as area graphs, forest plots, streamgraphs, or network analysis. All visualizations are available on the Wonderful Wednesday blog.

Read more...

For this edition of the Wonderful Wednesdays, the audience was asked to send examples of visualizations representing coronavirus data. Mark Baillie guided the webinar, showcasing different types of visualizations, such as area graphs, forest plots, streamgraphs, or network analysis. All visualizations are available on the Wonderful Wednesday blog.

The webinar starts by showing an example of a governmental COVID dashboard. Clean and straight to the point, the UK site provides information about the pandemic in the country. It includes numbers and texts that answer the main questions that users might have: are cases rising? What are the vaccinations rates? By clicking each of the topics, users can see specific information in the form of interactive charts. 

The next visualization, by the Financial Times, represents the excess of deaths caused by the Coronavirus pandemic. It shows a set of small multiples -one per country- highlighting when the number of deaths was higher than the average in recent years. Although the information is impressive, the chart is very crowded with data and annotations which makes understanding it a bit difficult. A similar project, done by The Economist, included interactivity to avoid this problem. Their implementation of tooltips to extract information is very well done.

The following visualization is an unusual chart done again by The Financial Times: a streamgraph. This type of plot is a stacked area chart displaced around a central axis. Its organic shapes and different looks might make them unconventional and attractive. However, with this type of chart is difficult to compare the size of the areas. And users who are not familiarized with them might have a hard time understanding them.

But visualizing the coronavirus pandemic doesn’t mean using rare charts, or interactive projects. One forest plot from The Economist shows how this traditional chart can be extremely successful at highlighting stories. The key is to include labels and annotations to make the information accessible and declutter the axis to make the data be the main protagonist.

Many of the projects around COVID used gamification and interactivity to explain complex scientific terms to the general audience. The webinar highlighted two that explained how the virus can expand slower or faster depending on people staying or not at home. Users can understand better how simulations and models work by interacting with the different variables that feed the interactive visualizations.

The last example is a network analysis of how the virus moved in Hong Kong during the different waves.
Page:

Upcoming Events

Latest Jobs