Digital Health

VR/AR is set to help life science companies uncover new data correlations, accelerating discovery and enabling more powerful predictions about the future. This technology could support regulatory and quality departments in visualising and acting on regulated product information and market intelligence

Romuald Braun at Amplexor

All businesses are swamped by data. If only they could get to the real insights more speedily, they could truly transform the way they operate. Colourful, graphical dashboards with drill-down detail have done much to bring data alive for decision makers, drawing their attention to what’s important and giving them chance to ‘slice and dice’ the data in a range of different ways to see what’s really going on. The logical next development is to be able to ‘walk through’ the data, uncovering new correlations and insights using virtual or augmented reality (VR/AR).

It might sound like something from science fiction, but, technically, much of this is already possible. So, how might it apply to life science and its management of regulated information and associated operational processes?

“Increasingly intuitive, 3D data modelling and visualisation will become the norm, bringing data alive in all kinds of important new ways for companies, and supported by technology that is becoming an evermore seamless part of how people work”

First, it’s worth delving deeper into the possibility that we might need new ways of visualising and navigating complex datasets. With the evolution of IT, companies are collecting more and more data and making decisions based on them. However, to make meaningful decisions, they need to turn all of these Big Data into information and experiences that their businesses can learn from and use as the basis for new action. Moore’s Law, the observation that the number of transistors in a dense integrated circuit doubles about every two years, could be applied to data too – volumes are doubling with small units of time, so that companies have more, potentially important, information than they know what to do with.

Until now, companies have relied on two-dimensional ways of looking at this, limiting the correlations that can be made, or the conclusions drawn. Insights inevitably depend on the angle at which the data are being looked at – when there may be more interesting and, so far, undiscovered insights within the data layers.

The ability to represent different data objects or assets in 3D models, and turn them in different ways to reveal different perspectives, could be transformational in delivering a richer understanding of a situation, and in projecting how this will play out under different parameters. There is an opportunity to uncover previously unseen patterns and trends, and combine them to see the correlations between diverse data sources. The extra dimension suddenly allows teams to ‘look around the corner’ of data and what they are telling them.

As humans, we are visual beings – so it follows that we will derive greater value from the ability to more intuitively ‘see’ what we’re looking at, as something more than sets of numbers in databases, tables, or bar charts. As volumes of data multiply, existing analytics and reporting tools are hitting the ceiling in terms of delivering clarity.

New ways of visualising and navigating data also present new scope for collaborating on data discovery, applying the principle that several brains are better than one. The more people who are looking at the data from different angles, the more teams will be inspired to dig deeper, or to bring further datasets into the analysis.

So, what, specifically, will virtual and augmented (a blend of virtual and physical worlds) reality bring to life science?

It’s probably true that 90% of the available VR apps are gaming-oriented today, but even these involve strategy building, and use experiences and data interdependencies to model and forecast future outcomes. In educational and scientific fields, there is a growing range of both free and paid-for VR tools available for investigating and exploring data. Add to this the immense computing capacity now readily available via the cloud, and these capabilities are becoming well within reach of non-IT people who want to expand their knowledge by looking deeper into data. The technology and its application are evolving all the time.

In life science, once you apply an object data model across regulatory and quality management processes – so that different data fields such as country, drug type, dossier, and document are represented and can be viewed in different combinations by their different interdependencies – there are numerous practical ways companies could apply VR- and ARbased data visualisation and navigation for useful effect.

Here are seven examples:

1. Pharmacovigilance and safety, for signal detection

Take the current situation with the COVID-19 vaccines, for which Phase III clinical trials are being conducted with people out in the real world, because of the urgent need to roll out the protection.

Although the vaccines have been authorised as being safe to use, mass monitoring for potential adverse effects is paramount, which means collecting huge volumes of data and analysing them in a comprehensive way. With five billion people being targeted, and each individual potentially generating 1Mb of data, that’s an unthinkably challenging prospect – overwhelming not just scientific brain capacity, but the scope of artificial intelligence (assuming this hasn’t yet been sufficiently trained in what to watch out for). Those responsible need to be able to represent and configure the data in different ways to spot and compare potential adverse effects.

2. Impact assessment, forecasting, and simulation

If there is a change in regulatory requirements, VR or AR visualisation offers a chance to ‘pull on that string’ and visually see how the impact of that change cascades through its operations and current assets.

Although it’s already possible to conduct fairly extensive impact assessments using software, the addition of a third dimension would allow teams to factor in the current availability of resources and of network infrastructure as part of the calculations, and weigh up all of the correlations simultaneously.

Once companies can visualise the impact of a change across all affected products, they can more accurately determine how realistic it will be to achieve that within the given timeframe.

3. Quality consistency checks across multiple data types and formats

Today, data sources take multiple different forms, from structured data and numerical values to free-form text, images, video, and audio files. Making sense of all of this, and being able to rely on what these diverse data are saying, means having confidence in the quality of these contributing sources – and being able to spot any overlap.

Introducing the VR/AR element could help ensure consistency across all data and metadata, highlighting anything that needs to be corrected, completed, or removed due to duplication.

4. Including emotion in pharmacovigilance reporting

If headaches are emerging as a common adverse event, whether linked to a COVID-19 treatment or some other medical intervention, the ability to include the dimension of emotion in analyses could help determine whether stress and anxiety might be significant contributors.

5. Clinical trials planning and management

Clinical studies can be harder to plan and recruit for, as pharmaceutical companies’ focus turns away from blockbuster drugs towards more specialised medicine, such as therapies for rare diseases. Adding a VR capability to clinical study planning and management, including dimensions for patient recruitment and availability, could make it easier to factor in all the variables and make more realistic calculations.

6. Manufacturing and distribution impact assessments, forecasts, and simulations to aid planning

Getting the Pfizer-BioNTech COVID-19 vaccine to market required a complex logistics chain and infrastructure, because of its particular temperature requirements. Being able to navigate the complex considerations visually across multidimensional datasets enables accurate planning, including any contingencies required.

Roads and trucks could represent supply and demand, and colours could signal time or quality issues. In the context of COVID-19, a model of the earth and spike lengths could signal where peaks of the virus are currently, or where demand is building, or least fulfilled.

7. Planning and managing marketing authorisation and variation submissions

Assessing the progress of electronic common technical document submissions by being able to visualise and navigate these as 3D pyramids, and see at a glance which parts are incomplete or waiting for documents or data, and which submissions have deadlines approaching, aided by colour coding, could make it much easier for regulatory teams to keep things moving.

As identification of medicinal products submissions become obligatory, advanced data visualisation could provide an invaluable overview across all the different data dimensions, helping companies cope with their increasingly complex data gathering and maintenance burden.

Once-cumbersome VR headsets are already giving way to slimmer glasses without wires – which, in turn, could be exchanged for content lenses. All of the heavy lifting, such as the intense data processing, can happen in the cloud today, connected via continuously-improving network bandwidth.

AR has its own distinct value in bringing digital and physical domains closer to performing real-time analyses. Take the scenario of a pharmacist asked to perform an inventory check to identify any non-compliant products in their stock room following a regulatory change. It’s possible to imagine scenarios where they would simply put on a headset that automatically scans for affected products, comparing the labels of products on the shelves with the correct latest information.

AR and VR technology promises more natural and collaborative engagement with data. It replaces the traditional one-dimensional, sequential view of data with a more real-world approach. Increasingly intuitive, 3D data modelling and visualisation will become the norm, bringing data alive in all kinds of important new ways for companies, and supported by technology that is becoming an evermore seamless part of how people work.