By Ruben Rodriguez (ERNI Spain)
When we look at the evolution of technology in the diagnostics area, we can differentiate three main phases in the evolution of its intended use. At the beginning, diagnostics companies were focused on improving analytic instruments in terms of accuracy, velocity and usage. In this first phase, the need appeared to centralise and manage the information provided by the instruments to improve medical decisions.
The second phase was focused in the automation of laboratories. Instruments were not isolated anymore. Step by step, the whole laboratory became a big interconnected system, and the sample workflows were centralised and controlled automatically. The focus was put on maximising performance and minimising human interaction.
Nowadays, we are finding ourselves in the first steps of the third phase in the area of diagnostics automation: use the technology for moving from sickness care to health care or, in other words, from being reactive to being proactive.
The power of data
Diagnostics companies are starting to discover the real potential of data processing and management.
Even with very different types of labs (in terms of sizing, automation level, etc.), here’s at least one thing all of them have in common: they are now starting to face the same kinds of challenges that other sectors, like manufacturing or logistics, have been used to dealing with for years in terms of equipment maintenance or performance. In addition, they are managing a huge amount of very valuable information on the people in the areas where they operate.
In an interconnected world, the diagnostics companies are seeing that the key for improving the service provided to their customers and patients, moving their focus from reaction to action, is to be able to apply the existing technology to get the proper data from instruments and laboratories, and manage it in the right way.
Here I’d like to talk about two relevant areas where this data management is becoming a differentiating factor for a company in front of their competitors:
1. Moving from reaction to prediction
The biggest trends for the future lead to global connectivity. The idea is to get data from everywhere every time, and then analyse, combine and learn from it to make better decisions.
It is well known how this is applied for commercials, but what happens if you apply it to health care?
The answer is clear: you can start engaging in prevention proactively, so you can move from “sickness care” to “real health care”.
Let’s use an example: if you get data from all EU countries about the level of glucose in its population and are able to combine it with population demographic data (age, sex, country, city, etc.), you may realise that, e.g., men in a certain part of the world have a higher probability for some kinds of diseases. In the end, that means you’ll be able to take preventive actions which are precise and effective to mitigate the risk.
Diagnostics companies have direct access to this kind of information. If they are able to manage this data, using IoT and cloud services to collect it, and data analysis and machine learning to process it, they’ll be able to provide this information, which could change the whole health system approach.
In the same way, if diagnostics companies are able to get data from their devices about their hardware components, they can apply the same principles and technologies for doing predictive maintenance of their devices.
This way, they can plan maintenance actions even before their customers realise they need them.
2. Simulations: looking for the optimal setup
The usage of digital twins is a technique already well proven in many industries, i.e. logistics or manufacturing. Nowadays, diagnostics companies are starting to introduce it to offer a better service to their customers.
The virtual simulation of the instruments allows the creation of virtual laboratories that are digital replicas of the real ones.
If we think of a laboratory as a factory of tests, the advantages of using digital twins are clear: Any change can be tested, any issue can be reproduced and a huge amount of data can be collected without the need of stopping them or even being there.
Testing everything with the greatest optimisation and then implementing it in the real lab decreases time for testing by 80 percent and significantly reduces the related costs.
Also, the data provided by the simulation can be processed by means of machine learning algorithms and used to try an endless number of configurations and setups, analyse them and check how the change influenced the sample management, the system throughput and the workflow performance.
In summary, technologies like Data Analysis, IoT, Machine Learning, Data Management or Cloud are becoming key in Diagnostics & MedTech companies. At the end of the day, these are also the main ones we are working with at ERNI in Spain.