While the exact definition and parameters of the lab of the future hasn’t been determined, it is clear that with technology and the wider environment continuing to evolve and innovate at greater speeds, there’s still plenty of scope for improvement in laboratories. The lab of the future - an innovative visualisation of laboratories with simplified workflows and smart technology at the forefront - has been widely discussed among researchers in the past few years.
In particular, two technology advancements have been getting more attention than the rest - AI and IoT. Being complementary fields, they not only work well together to improve the processes in a lab, but can also enhance the capabilities of one another and most importantly help with the massive amounts of data generated while undertaking scientific research. As well as technological advancements, the wider implications of the future of science and what it means for sustainability must also be discussed to gain a better understanding of how labs can strive to improve their processes.
In this blog, we will discuss the main points of a laboratory of the future, diving deeper into the aforementioned fields.
Digitalisation and IoT
A connected lab with data flowing smoothly between the equipment, data storage and the end user - while that might sound like a dream, IoT turns this into reality. From reservations and planning controlled access, to monitoring equipment remotely and alerting users promptly whenever there are any irregularities, IoT can help both accelerate science and ease the workload of researchers.
While it is clear that both technological and scientific advancements usually go hand-in-hand, looking into the progression it is apparent that scientific advancements help accelerate and improve scientific processes thus leading to speedier discoveries. Even though many labs are still far from perfect, the most pressing issue to be addressed is saving researchers’ time - and this is where IoT solutions step in. Being described in the most simple terms as ‘all devices connected to the internet’, IoT can take many different forms in the laboratory, and the most common is in the form of automation. IoT can also help keep lab activities on track and reduce admin work such as checking if equipment needs maintenance, taking inventory of supplies and much more. In addition to that, IoT devices can help monitor experiments in real-time remotely thus making the devices much ‘smarter’ and saving researchers a lot of time. By monitoring the experiments, IoT devices can also alert researchers at given times when an event has been triggered and help lower human error as well as be an efficient way to collect data and conduct experiments through providing support with automatic data capture and analysis.
While IoT might seem like the perfect solution to bring us closer to the lab of the future, at the moment there are some drawbacks to be addressed as well - high costs and compatibility being the main ones. While in the ideal scenario all of the instruments within a laboratory would be connected to an IoT device, in reality many of the pieces are not IoT compatible yet. Overcoming the compatibility issues would speed up the development of the laboratory itself thus also improving the way researchers work.
Automation, AI & machine learning
While the automation trend is continuously growing, one benefit of utilising automation in routine experiments can be the significantly increased accuracy of findings and reduction of the time and manual work that is usually required. Scientists should see a reduction in the time they have to spend performing analysis of experiments through being able to use the now-centralised data with various data science and analytics tools such as those provided by Tetrascience, and designing other experiments instead. A strong focus on automation as well as the use of AI in research can also help improve the process of analysing data and results.
To further develop the knowledge of automation and AI in science, we hosted the 4th edition of our Lab Talks panel discussion, where we debated the opportunities and drawbacks of implementing automation in science research with professionals working in the field - Joana Rocha - a bioengineer & research fellow at INESC Technology and Science, Fane Mensah - life science business director at Synthace, and Kimberly Holtz - sales and business development manager at IRIS.AI.
One benefit of automation that arose during the discussion was the chance to “provide the physicians with a second opinion” as Rocha suggested, which helps with her work of automating diagnoses. Indeed, adding a second opinion that isn’t affected by potential human error can be largely beneficial when the result isn’t always clear to the physician. Additionally, the efficiency of automating research in literature was said to be a great advantage for scientists by Holtz, which can be crucial when dealing with widely researched topics that have a “huge amount of scientific literature” available. As well as the enhanced efficiency in research, the complexity of the work can also be increased when machine learning and automation is implemented, leading to “groundbreaking work”, especially from scientists that are young or from different fields, Mensah added.
However, with the benefits of automation and AI in science research, there may also be drawbacks or considerations that need to be discussed in order to prevent any issues from arising during the research. One of these issues is the ethical implications and possible bias behind machine learning- when training an algorithm for example, the sample must be diverse enough to encompass the population. Rocha suggested that when implementing speech recognition if there is only one language and dialect used, the result would exclude other members of the group that speak in different ways or languages. This would mean that any results gained from such an algorithm would not only be discriminatory, but also not representative or reliable.
Another issue that may impact research negatively may be the expectations that involving AI can incur- Holtz explained that even though AI can simplify and enhance scientific work, the researcher should still “stay in control of what the machine is doing” and double check the results that are produced, as there are still limitations to what AI can do within research.
Green labs and sustainability
While thinking about the lab of the future may conjure up images of new technology and advancements, another important factor that must be considered is sustainability and how research can be carried out without harming the environment or wasting resources.
It is especially important to address this due to the fact that scientific labs consume a high amount of energy and can also be a barrier to sustainability as green initiatives can be challenging to implement without sacrificing good laboratory practices.
One way in which labs can become more ‘green’ is by adopting simple practices such as energy meters or timers on large appliances, turning off equipment when not in use, and performing regular maintenance on equipment to keep it performing optimally. These small steps could reduce energy consumption significantly, and also reduce bills for laboratories.
Unfortunately, it can be common for researchers to avoid implementing sustainability measures because of the aforementioned laboratory practices. Due to safety issues, it may be assumed that the amount of waste produced by a lab cannot be reduced but this is not the case. For example, investing in glass containers as opposed to utilising single use plastic for certain instruments is a good way to help eliminate waste in the lab. Additionally, choosing more sustainable product options such as suppliers that provide less packaging or are based more locally to reduce emissions from travel are other ways in which it’s possible to reduce waste and help the environment.
With the advancements mentioned, achieving a sustainable, efficient and all-over digital lab of the future is something we could be looking into in the near future.