The Lab of the Future: Making the shift from reactive to predictive labs
Being at the forefront of worldwide R&D IT innovation as a member of the Pistoia Alliance is quite exciting. Along with fellow members like AstraZeneca, Merck, Pfizer, Dotmatics and The Hyve, Vivenics contributes to lowering barriers to innovation in R&D. Introducing Industry 4.0 in lab environments is one of the many ways to achieve this. Just imagine what technologies such as Artificial Intelligence, Machine/Deep Learning and Internet of (Lab) Things can contribute to labs in the long run!
Taking part in Pistoia Alliance’s ‘Lab of the future’ working group is a journey into the Digital Lab of the Future (LotF). A future that is in fact a lot more imminent than many people think, and that has the potential to revolutionize health care in general and R&D environments in particular. We are already witnessing a major shift from ‘reactive’ to ‘predictive’ labs. Labs will increasingly unite three worlds: people (scientists, patients, health professionals), businesses (including lab automation / lab information systems and analytics) and Internet of Things (IoT). As these elements become more and more interwoven, the outcome becomes increasingly predictive, especially when driven by AI, Machine Learning and Business Intelligence.
Internet of Lab Things
According to many experts, the Internet of Things is likely to reach its maturity in the next couple of years. Compared to industries such as transportation, consumer markets and security/surveillance, R&D labs have some serious catching up to do in this field. Accelerating the Internet of Lab Things will, among other things, enhance digital data capture, increase flexibility and provide labs with real-time data. Within Pistoia’s Lab of the Future Community of Interest, a project on predictive maintenance of freezers has been initiated. Some time ago, Shire and TetraScience started installing IoT sensors on freezers. Going forward, data from these sensors at several member companies will be collected and stored centrally and will be disclosed to all PA members. AI and Machine Learning technology is then used to discover possible patterns.
These analyses will eventually allow pharmaceutical companies to predict when a particular freezer needs to be serviced or replaced before any disruptions occur and biological material becomes unusable. Based on these findings, the project will be extended to other instruments including balances (drift and calibration), temperature gauges and humidity meters. At the same time, switching to a framework based on the Internet of Things (IoT) could remove data roadblocks in multi-vendor setups, which currently slow down workflow. Consortia like the not-for-profit Standardization in Laboratory Automation (SiLA) initiative and Allotrope Foundation are bringing together pioneers from this field, easing collaboration between instrument vendors and promoting open source and exchange of instrument protocols. By bridging the IoT gap, the pharmaceutical industry will bring new and very promising opportunities within reach.
Smartwatches and ‘nanobots’
Many of us are already monitoring our heart rate and counting our daily steps with a smartwatch. In a few decades we should not be surprised if so-called nanobots, or mini-robots, are injected into our bloodstream to look for malicious intruders like viruses and transmit the results to a doctor in real-time. Sooner rather than later, we may rely more on data stored by cloud providers like Amazon, Microsoft Azure and Google than on our own doctor. After all, based on big data a prediction can be made as to when someone will become sick even before the actual symptoms emerge. Worries about the protection of this personal data will quite likely be addressed using blockchain. This technology, which is explained elsewhere on our website, can be seen as a kind of distributed database. It has very sophisticated traceability mechanisms and is already in use to protect cryptocurrencies such as bitcoins against fraud or manipulation.
Standardization of data formats
Meanwhile, other topics such as the standardization of data formats to safeguard retrieval of data over decades, as initiated by the Allotrope Foundation, will have to remain on the mutual agenda as well. The same applies to enhancing laboratory information management systems (LIMS) and electronic laboratory notebooks (ELNs) which are unlocking the potential for new cloud-based sciences, including remote execution of experiments. SiLA is also promoting AnIML, a basic-level data format that counters the problem of lab instruments all speaking a different language. Aligning the communication medium between manufacturers using an XML-based data format would not only solve this problem but would, at the same time, allow potential off-the-shelf use of thousands of new tools. Last but not least, the use of visualization and data analysis software like Vortex and KNIME is indispensable in keeping pace with the data explosion that is already taking place in life sciences R&D.
In parallel, hardware such as large format screens, touch-walls, and high quality video and web conferencing will increasingly prove its value in accelerating analysis and promoting collaboration. Whoever develops faster and more accurate automated analytical software will have a competitive advantage, ultimately reaching the patent filing line first. Many projects that will follow in the near future, initiated either by Pistoia Alliance or others, will continue to be about sharing information and experience as well as sharing costs for the evaluation of new technologies by individual members. Moreover, it allows for jointly identifying critical challenges and providing information about maturity and ROI as well as influencing stakeholders towards aligned approaches in terms of standards and best practices. From our perspective, it is already difficult to be patient when it comes to the Lab of the Future, let alone be ‘a’ patient waiting for a cure.
Introducing ‘Industry 4.0’ in lab environments will boost innovation in R&D