16 - AI: Ahead in the clouds

Barely a week goes by without more exciting reports of life sciences companies pushing the boundaries of what can be achieved using AI. But what defines success often comes down to the marriage of the right data with the right analytics.

The potential exists to transform everything from drug discovery and design, to clinical trials and commercial operations. Indeed, we’re starting to see real-world examples of how advanced analytic capabilities accelerate new therapeutic treatments.

In one partnership between AstraZeneca and the University of Sheffield, AI is deployed to more accurately predict whether a drug will successfully target a cancer-related protein, or impact other cellular targets that cause side effects. It’s just one example of the interplay between life sciences and computer sciences in solving some of the biggest challenges in modern medicine.

AI and cloud analytics are also bringing us closer to personalised medicines being rolled out across our healthcare systems. This could drastically improve patient outcomes and help healthcare providers optimise their resources and costs.

Even the use of patient genomic data is helping to refine how the best medicines for the right patients are matched together. Targeting treatments in this way brings commercial benefits for life sciences companies too, since they no longer need to wait decades to release the next blockbuster.

It could also remove some of the limitations – and ethical implications – of using placebos in clinical trials. For instance, Nino da Silva of BC Platforms, a technology company specialising in personalised medicine, has highlighted the growing importance of being able to integrate large amounts of genetic data, while establishing synthetic data control arms to enable drug approval in different territories without recruiting more volunteers.

This is now possible because researchers are able to use AI analytics to rapidly make sense of vast amounts of data with more granularity to support predictive decision-making. With new AI models being continually developed and deployed all the time, there is almost no limit in the amount or complexity of data that can be processed into actionable decision points.

Since the COVID-19 breakout, life sciences companies increasingly recognise that digital transformation in research and development, including the application of AI and advanced analytics, is no longer a ‘nice to have’ but a necessity in today’s world.

AI-powered predictive analytics can streamline the route from product idea to regulatory submission and inform operational decisions around drug development and design. The goal is to drive innovation in a highly regulated sector, while providing life-saving and life-enhancing treatments.

Of course, the right level of data governance – including organisational-wide checks and processes for data quality, and human oversight and accountability – is critical if the technology is to deliver on its promise. Indeed, predictive algorithmic models that are continually monitored and revised as new data is fed in reduce the risk of human biases adversely impacting decisions.

Secure cloud platforms enable multidisciplinary teams to develop, manage and enable modern ‘statistical compute environments’, where advanced analytics tools such as AI can be applied to both proprietary and third-party data. With this, researchers are able to conduct clinical trials faster and more efficiently.

Ahead in the clouds

Analyse this

Analyse that