We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.

Advertisement

How Will Automation and Digital Technology Shape the Lab of the Future?

Data and AI symbols floating above a laptop.
Credit: iStock
Listen with
Speechify
0:00
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 9 minutes

Advances in automation and digitization are driving the journey to the lab of the future, a vision of a lab where researchers are freed from manual repetitive tasks and efficiency and innovation are at the forefront.


Some experts have estimated that 70% of lab workers' time can be wasted on administrative tasks, data analysis and reporting. In the lab of the future, a marriage between artificial intelligence (AI) and robotics will free up time for researchers to focus on more important tasks, such as interpreting data and answering scientific questions.


The development of smart labs that utilize the Internet of Things (IoT) to improve data connectivity and autonomous robotic systems that can better navigate a lab environment are just some examples of the technologies restructuring how research is conducted.


This article explores these advances further, discussing the roadblocks preventing labs from utilizing these transformative technologies and evaluating how digitization and automation will continue to shape the research landscape.

AI-powered robots pave the way for lab automation

Laboratory automation is designed to eliminate the most repetitive tasks in a lab, such as liquid handling. As technology has progressed more complex processes can now be automated, including entire workflows. Lab automation is usually classified as either partial laboratory automation (the automation of a single step in a lab process) or total laboratory automation (complete automation of a lab process or workflow).


Driving lab automation is the progress made in robotics enabling labs to streamline more complex manual processes.


Dr. Gabriella Pizzuto is a lecturer in robotics and chemistry automation at the University of Liverpool. Her research at the forefront of artificial intelligence and robotics is focused on improving current robotic systems in labs to make them smarter, more efficient and safer when working in human-centric environments.


One of the projects Pizzuto has been involved in is advancing modular, multi-robot integration in laboratories.1  “There are many advantages to laboratory robots, for example, being able to work around the clock, carry out longer experiments and tackle reproducibility challenges,” Pizzuto told Technology Networks.


A survey of 1,500 scientists by Nature found that more than 70% of researchers have tried and failed to reproduce another scientist's experiments.2 Many hope that lab robotics will help overcome the reproducibility crisis plaguing scientific research. “With a robot, you have more of a controlled experiment, you remove elements of human and repetitive error. Plus, there is a lot of data that you can record and gather from a robot,” said Pizzuto.


Before the potential of laboratory robots can be fully realized safety and cost concerns must be overcome, Pizzuto explained, “You would need to make sure the robot is as safe as possible. For example, if a robot working 24/7 drops something overnight, the researcher might be exposed to whatever the robot has dropped when they next enter the lab.”


“Robots to this day are still quite expensive and so not accessible to everyone,” continued Pizzuto. “Historically, lab-based scientists won’t have experience using robotics so there are also training barriers that need to be overcome before robots can become commonplace in labs.”


Traditional laboratory automation focused heavily on rigid machines specific to a single task. The next generation of lab robots should be flexible and capable of carrying out a variety of laboratory tasks.


“Researchers are now more interested in intelligent robotics, or autonomous robotic systems. For example, if a robotic system has been developed and deployed for nuclear applications, there could be knowledge transfer to our laboratory environments,” said Pizzuto.



Advertisement

Recent advances in robotics have also focused on incorporating AI and machine learning to improve current “robotic scientists”.  A “self-driving” laboratory of robotic equipment powered by a simple machine learning model has successfully reengineered enzymes to be more tolerable to heat without human input save for hardware fixes.3


“In our lab, we use machine learning methods, including supervised learning, reinforcement learning and more recently generative AI methods to teach our robots different skills and tasks within laboratory experiments,” said Pizzuto. “This is to improve how the robots interact with laboratory environments.”

Enter the digital lab

As labs continue to incorporate digital hardware, it is more common to see researchers interacting with technology daily. This digital transformation has highlighted the importance of connectivity and security in laboratory practice.


“If it's done well, there are a lot of pros both from an academic and an industrial perspective for increasing digitization in the lab,” Dr. Samantha Pearman-Kanza, senior enterprise fellow at the University of Southampton told Technology Networks.


Pearman-Kanza investigates the social aspects of adopting digital technology to understand the barriers that stop labs from going digital. Her research involves applying technologies like electronic lab notebooks (ELNs) and IoT devices to improve the digitization and knowledge management of the scientific record.


“Data that is preserved digitally is more secure. People don't put much stock in backing up physical paperwork. By putting something on a computer, you're suddenly affording it more importance,” explained Pearman-Kanza. “Additionally, integrating data on a single platform encourages collaboration within and between labs.”


ELNs and laboratory information management systems (LIMS) are vital to ensuring lab data is FAIR (findable, accessible, interoperable and reusable).4 Both of these pieces of software are designed to improve the traceability of lab data and facilitate compliance and data integrity. Cloud-based solutions are available that promise built-in security and regulatory compliance. These third-party solutions reduce the need for traditional IT infrastructure, such as servers and storage devices.


Building on data connectivity, smart labs have been created that utilize IoT technologies to improve instrument inter-connectedness. A demo smart lab has been set up at the University of Southampton called Talk2Lab to investigate how IoT devices can be integrated into the lab. One technology they have tested is voice command systems.


“Integrating voice allows you to ask questions while you're completing other tasks in the lab. We have tried commands such as: can you show me the dashboard for this, what's the temperature of this,” said Pearman-Kanza. “In the future, it could be possible to integrate this with ELNs to take notes using your voice alone.”

“For me, the grand vision of the smart lab would be to have everything interconnected with all the instruments ‘talking’ to each other,” said Pearman-Kanza.

Another key aspect of a smart lab is the integration of smart cameras and sensors. The smart lab setup at the University of Southampton uses cameras that have alleviated the need for an additional staff member to be present during the calibration of certain instruments.


The smart lab at the University of Southampton has also incorporated sensors that measure laser power. “Often our laser can short out and this can ruin experiments being set up,” said Pearman-Kanza. “Using sensors, we're looking at how we can take the data on laser power and make predictions on when we think that's going to happen, to save people time with their experiments.”


Digital solutions of the future will need to address individual laboratories' pain points and needs. To maximize the benefits of digitization, labs need a clear vision of how to use technology in a way that maximizes their long-term “digital health”.

Building trust in artificial intelligence

Artificial intelligence is continuing to become more widespread in scientific research. A recent Elsevier survey of corporate R&D professionals found that 96% of researchers think AI will accelerate knowledge discovery, while 71% say the impact of AI in their area will be transformative or significant.

Figure 1: Results from a 2023 survey by the nonprofit Pistoia Alliance showing the lab technologies that life science companies plan to invest in over the next two years. Credit: Technology Networks.


Despite positive sentiments, the report also highlighted researchers' concerns about ethics, transparency and accuracy. There is particular concern over the use of AI in science publishing, with various examples of non-disclosed AI-generated text labels having already appeared in peer-reviewed research from different journals.


As developers work through these issues, researchers are finding unique ways to utilize AI to help improve scientific data handling. These applications include automating data collection and curation and optimizing drug discovery and clinical trial design processes.5


The next frontier in bringing AI into labs will likely involve overcoming researchers' fears of the technology. “When there is a new technology that people don't completely understand, there will always be a level of fear,” said Pearman-Kanza. “If we can provide transparency and accountability, this will go a long way to increasing trust in AI technologies.”

Overcoming digitization and automaton roadblocks

Many labs today don’t lend themselves to the modern requirements of automation and digitization. Assessing whether a lab has the required hardware, space, power supplies and internet connection are just some initial factors to consider when onboarding digital and automation technologies.


“You need dedicated lab hardware,” explained Pearman-Kanza. “One day in the future, there may exist self-cleaning laptops or ways to sterilize all the chemicals off hardware in a way that isn't damaging. But until we get to that, you don’t want to bring contaminated hardware in and out of the lab.”


Cost and a lack of specialist knowledge also continue to hamper efforts to onboard these technologies in labs. Pizzuto stated, “Robots are a significant additional cost that labs would have to weigh up.”

“We need to promote more teams where engineers, computer scientists and lab-based scientists such as chemists and biologists come together to address the current problems with automation in the lab to make these technologies more accessible,” said Pizzuto.

An evolving landscape

Digital technology and automation are becoming more prevalent in labs, Pizzuto imagines that within the next few years “Most labs, especially the newer labs, would be using some form of AI within their research and experiment.”


The adoption of robotics may be slower than AI due to cost constraints, but Pizzuto envisions the use of robotic arms, dual-arm robotic systems and mobile manipulators increasing to help with labor-intensive workflows. Mobile manipulators that can transport samples between different stations will play an important role in connecting workstations and different labs within the same building.


“When we think about using more robotics and AI, data will become crucial,” said Pizzuto. “Having this data stored securely on a cloud-based software and having models on the cloud that can analyze these large volumes of data, will be vital.”


Technologies used in other aspects of our day-to-day lives will continue to find unique uses within the lab. Virtual reality (VR) headsets are now widely used by gamers for a more immersive experience. These smart wearables are also being utilized in research, for example, to help navigate microscopy data.6 Northwestern University researchers have developed VR goggles for lab mice to simulate natural environments to more accurately and precisely study the neural circuitry that underlies behavior.7


“I expect to see the amount of digital hardware in the lab continuing to increase,” said Pearman-Kanza. “Many existing technologies we use in our daily life will find applications in the lab. Hardware you can interact with that also has computing power such as tablets will keep increasing in prevalence.”


The future of lab automation and digitization will be a combined human and machine effort requiring new expertise and dedicated data stewards. Pearman-Kanza concluded, “I think it’s important to remember that one of the common goals with technology is to make data more reliable and reproducible. A successful digital future will involve understanding where we need people and where we can best use technology to optimize human efforts.”


About the interviewees:


Dr. Gabriella Pizzuto is a lecturer in robotics and chemistry automation and a Royal Academy of Engineering research fellow at the University of Liverpool, and a research area lead in chemical materials design at the Henry Royce Institute. Pizzuto obtained her PhD in computer science from the University of Manchester. Her research focuses on developing new methods for creating the next generation of robotic scientists.


Dr. Samantha Pearman-Kanza is a senior enterprise fellow at the University of Southampton. Pearman-Kanza obtained her PhD in web science from the University of Southampton. Her research involves applying computer science techniques to the scientific domain, specifically through the use of semantic web technologies and artificial intelligence.


References

1. Lunt AM, Fakhruldeen H, Pizzuto G, et al. Modular, multi-robot integration of laboratories: an autonomous workflow for solid-state chemistry. Chem Sci. 2024;15(7):2456-2463. doi: 10.1039/D3SC06206F

2. Baker M. 1,500 scientists lift the lid on reproducibility. Nature. 2016;533(7604):452-454. doi: 10.1038/533452a

3. Rapp JT, Bremer BJ, Romero PA. Self-driving laboratories to autonomously navigate the protein fitness landscape. Nat Chem Eng. 2024;1(1):97-107. doi: 10.1038/s44286-023-00002-4

4. Wilkinson MD, Dumontier M, Aalbersberg IjJ, et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data. 2016;3(1):160018. doi: 10.1038/sdata.2016.18

5. Wang H, Fu T, Du Y, et al. Scientific discovery in the age of artificial intelligence. Nature. 2023;620(7972):47-60. doi: 10.1038/s41586-023-06221-2

6. Spark A, Kitching A, Esteban-Ferrer D, et al. vLUME: 3D virtual reality for single-molecule localization microscopy. Nat Methods. 2020;17(11):1097-1099. doi: 10.1038/s41592-020-0962-1

7. Xia M, Ma J, Wu M, et al. Generation of innervated cochlear organoid recapitulates early development of auditory unit. Stem Cell Rep. 2023;18(1):319-336. doi: 10.1016/j.stemcr.2022.11.024