Healthcare’s biggest problem today is that costs are being compounded by ageing populations. New sources of data and innovative technologies can help labs address this problem by giving them new insights and capabilities for patient care.  

As sensors become more sensitive, we will have an unprecedented amount of data to understand our health trajectories. Increasingly, we will use technology to diagnose, predict and prevent disease by analysing data from voice algorithms, facial analytics, social media and other novel sources. With a wealth of different data, and cheaper ways to process it, laboratories will play an ever-greater role in the shift towards preventative and personalised care.

Three core technology laws drive this data revolution. Moore’s law states that processing power doubles every 18 months at the same cost, Kryder’s law says the same about storage, and Metcalfe’s law – the law of networks – states that the value of a network grows as the square of its participants.

We’re starting to see a confluence of these three laws in healthcare. A clear example is the cost of processing a genome, which has dropped from $2.7 billion in 2001 to just $200 in 2018 [1]. Some predict that by the mid-2020s it may drop to less than 5 cents [2] – less than the cost of flushing a toilet. If that happens, sequencing services will become ubiquitous, facilitating preventative medicine at massive scale.

These trends are also enabling advances in artificial intelligence (AI) and automation. In one Chinese factory, 90% of human workers have been replaced [3] by automation. Production is up over 160%, while the defect rate dropped from 25% to 5%. If you’re a patient in a hospital, wouldn’t you want a similar reduction in defect rates?

AI algorithms and robotics are already used in healthcare, and robots can perform certain tasks better than humans. For some types of surgery, robots are more reliable than humans, and in some diagnostic tests, machine vision can pick up on details that humans cannot. AI is already being use by radiologists to diagnose an array of conditions, such as diseases of the heart, liver and bone. More applications are almost certainly on the way.

These advances are helping to improve access to quality healthcare by facilitating early detection of diseases. Early detection sensors for breast cancer [4], for example, are now approaching the accuracy of mammograms and will eventually be allow household detection and early intervention.

As we start to think about applying these sorts of technologies to our laboratories and businesses, we need to think of AI as an amplifier of human beings. The future of healthcare may be digital, but the reality is that AI cannot replace people. Our mindset should be about augmentation, not replacement.

For labs, it will be crucial to consider how AI can be used to reduce the cost of diagnostics, automate workflows and improve operating efficiencies, as well as how to generate new clinical insights from unprecedented amounts of data. This process will require creative thinking and experimentation. If successful, it will help drive greater access to healthcare services and reduce costs for patients everywhere.

[1] Now You Can Sequence Your Whole Genome for Just $200, Wired, 2018

[2] Soon, It Will Cost Less To Sequence A Genome Than To Flush A Toilet — And That Will Change Medicine Forever, Business Insider, 2014

[3] Man versus machine: 9 human jobs that have been taken over by robots, BT, 2018

[4] Laser-sonic scanner aims to replace mammograms for finding breast cancer, Science Daily, 2018

This article is based on the presentation: Healthcare innovations from remediation to prevention at the Roche Efficiency Days (RED) 2018 REDefining perspective in Guangzhou, China.