Data analytics in healthcare

Al Sisto

Blog by: Al Sisto - 13 / Aug / 2021

Most us have become far more familiar with data analytics in healthcare than we perhaps ever expected to over the past 18 months. We have pored over graphs of infection and hospitalisation rates – and their future predictions. We have marvelled at the science underpinning superfast vaccine rollouts – and read article after article discussing those vaccines’ efficiencies and side-effects. We have listened to scientists and analysts explaining why particular interventions are effective – or otherwise – and even attempting to rebuke misinformation about those same interventions.

This highly visible flurry of data analytics in healthcare is happening alongside a quiet – but much broader – data analytics revolution. As the healthcare sector is increasingly digitised, it is truly revolutionising how healthcare is organised, developed and delivered – with enormous potential for health and wellbeing.

The Internet of Things (IoT) is a vital part of the picture. As it becomes cheaper and easier to embed connected sensors in an array of different devices, so it becomes possible to collect healthcare information from an extraordinarily wide array of sources. Wearable devices, for example, can collect key health metrics such as heart rate, exercise volumes and calories burned, temperature and so on – on both a regular and an ongoing basis. From there, data analytics can operate both on an individual level, building an in-depth picture of the individual’s health over time – and on a group level – combining that individual’s data with information from hundreds, thousands or even millions of others to track overarching health and activity trends.

From there, the potential interventions simultaneously span a spectrum from micro to macro. On the individual level, for example, those who are failing to hit ideal levels of activity can be gently – or otherwise – encouraged or reminded to take more exercise. On a mass population level, targeted interventions to improve activity levels can be developed, for example by recognising the scenarios in which people are most likely to be active.

And these examples, in turn, are just one fraction of the overall healthcare IoT picture. More advanced sensors are not merely part of wearable devices worn on the body, but embedded devices implanted in the body. These may measure metrics like blood sugar level, for example. Here, as above, data analytics work on both an individual and a mass population level – but such devices can also frequently receive instructions back from the analytics platform. If blood sugar reaches a certain level, for example, an implanted device may be instructed to provide a shot of insulin. If heartrate is interrupted, a connected pacemaker may deliver an electrical pulse, and so on.

Of course, myriad dimensions of the healthcare sector do not directly ‘touch’ patients, but rather are to do with vast back office, resourcing and research functions. Here, again, the implications of IoT technology combined with big data analytics are enormous. Consider a typical large hospital, and the huge number of decisions which need to be made on a daily basis simply in terms of resourcing – not just staff, but also physical and energy resources. Data analytics can streamline and automate a great many of those decisions, particularly when combined with IoT technology in, for example, smart energy and lighting systems.

The healthcare sector has been a shining light of innovation, efficiency and compassion throughout the Covid-19 pandemic. As we emerge and enter a new era of healthcare, data analytics will be a critical role in ensuring that developments in diagnostics, treatment and delivery continue to be balanced with a truly human touch.


Topics: IoT, IIoT, healthcare, data analytics, Smart Healthcare

Example CTA

Subscribe to Email Updates

Posts by Topic

see all