Health-related data and its collection have grown exponentially recently, thanks in part to a range of digital tools and the Internet of Things (IoT) which uses sensors to collect millions of data points of physiological data from patients over a period of time. According to Promit Roy, PhD candidate, Trinity Business School, Trinity College, and associate director innovation, Novartis Ireland, the "enormous data these technologies and channels cater are very resource and time intensive."
Analysing this data and identifying patterns and signals and adverse events are “crucial for life science industries,” he says. “However, screening and processing data to eliminate noise and find knowledge is a challenge for the pharmaceutical industry. This requires new approaches to handle/screen them for useful information.” As a result, the sector is turning towards artificial intelligence (AI) to support this data collection and analysis.
Human vs machine
Martin Kussmann, chief scientific officer, Nuritas, explains how AI outperforms human intelligence. "Human intelligence is still unmatched when it comes to versatility: the human brain is extremely flexible in developing and executing a vast array of cognitive and creative capabilities.
“However, AI is increasingly outperforming human intelligence when it comes to data processing speed, handling huge volumes of data, establishing connections between [especially at a first glance unrelated] large sets of information, fast learning, and – first and foremost – forecasting and predicting scenarios, behaviour, and functions of complex systems.”
AI-inspired innovations
Dr Kussman explains that AI is already revolutionising the sector. “It has tremendous impact on medicine, nutrition, diagnostics, environmental science, logistics, robotics, and this impact will exponentially grow. Rational drug design for example is nowadays greatly enhanced by AI and yields more efficacious drugs in a shorter time.
“In medical diagnostics, AI’s superb capability of pattern recognition is widely used to improve and accelerate early and precise detection of disease states and deviations from healthy physiology, especially when combined with imaging technologies.”
Yufei Huang, associate professor in operations management, Trinity Business School, Trinity College, says that AI is an “enabler” for the life science sector and there are many ways it can be used.
These include: centralised clinical trials for new drugs and the ability to implement the trial remotely to participants across a wider geographical area, reducing the cost and increasing the change of relevant and accurate data; assisting with healthcare access and basic diagnostics from telemedicine in areas where there is a lack of trained professionals; and adverse event (AE) detection and management where real-time monitoring by AI allows patients to report side effects of drugs and reduces the overall time to identify and report AE cases to the life science company and health authorities.
Dr Huang says that due to a rising interest in body wearable sensors and their commercial availability that generating varied data from variable sources will soon be the new norm but that “it is our social, ethical and corporate responsibility to ensure that what we substitute for a human is validated enough to provide quality healthcare”