Leveraging AI in Cell Culture Analysis

Mammalian cell culture is a fundamental tool for many discoveries, innovations and products in the life sciences. Currently, cells are the smallest unit of sustainable life outside the body, thereby providing an essential platform for testing hypotheses and mimicking biological processes. The applications of cell culture, while not limitless, are plentiful. 

Every cell type, downstream application and use case is somewhat unique. With each new application, a laboratory team must learn, sometimes laboriously, the characteristics of the cell type and redefine optimal growth conditions or purchase the rights of a proprietary method, if it exists. The protein-producing machinery of some cell types can be harnessed to generate large quantities of therapeutic proteins. The complex, diseased state of primary tumor cells might be used to evaluate the response to drug compounds and perhaps even find a unique biomarker “signature” to explain an overly positive or negative result. Some immortalized cells are stored for later use in a quality control assay. Other cell types may be used to quantify the genomic, transcriptomic, and proteomic in response to external stimuli. 

During cell culture analysis, each time a visual inspection is performed under the microscope, observations and judgments are made about the cell and culture as a whole. Are there contaminants in the media? Are the cells stressed, perhaps as indicated by granules? What is the density of the cells in the culture dish (e.g. are they overcrowded)? What is the morphology of the cell, and is it consistent with expectations?  Is it ready for the next step in the protocol? 

Any of these visual observations can affect the cells and therefore the experiment. The uniqueness of each cell type and its associated experiment requires that the judgments be consistent and repeatable for optimal results. While a myriad of hardware exists to automate the physical manipulation processes, the reliance on visual observations means each operator requires extensive hands-on training from experienced senior operators working with that particular cell in order to make correct judgements and decisions. This heavy reliance on senior expertise can create bottlenecks and limitations to discovery. 

For companies focused on one application, senior operators can easily train, and scale, their team for a specific cell type and experiment. That team will analyze the same cell type day in and day out, over time developing expertise and intuition. But what about companies exploring multiple cell types, or businesses looking to add a new capability? Operators must learn the new cells and process, then train the team, creating additional strain on their time and limiting innovation.

How automated decision-making can streamline processes and assist operators with cell culture analysis

With artificial intelligence and machine learning, labs can now employ automated decision-making tools that can not just support cell culture analysis but streamline it as well. Instead of a largely manual, human-driven process, a customized tool would allow the organization to evolve with their cells efficiently and effectively, and in a way that different methods can be easily applied across cell types.

In this new environment, AI and machine learning tools can reduce the burden on human operators, material waste, and the subjectivity of the process to work towards laboratory automation. The process becomes more scalable, extending the capabilities of the team. Instead of an operator learning and maintaining ten different cell types, each one of them being a little different from the last, that same operator now has this machine learning tool acting as their eyes and ears. Instead of senior operators spending weeks or even months training junior operators on a specific cell type, trained models are available and usable by anyone in the laboratory, even the greenest operators. And if an experienced operator departs, taking years of knowledge and expertise with them, the established models ensure there is no impact on the quality and consistency of the established process. The same models can be used as a starting point for new cell types and studies. 

Cell culture has the potential to support innovation in countless ways, and the application of AI to cell culture analysis can offer meaningful change to both the operators and the business itself. Contact our team of experts today to see how we can support your scientific discoveries and advance your analysis processes.

About the Author

Robyn Cardwell, Ph.D. holds a Ph.D. in Biomedical Engineering from Virginia Tech & Wake Forest University, and a B.S. in Mechanical Engineering from Colorado State University. She is a biomedical engineer focused on the intersection of biological processes with engineering principles. Robyn previously spent many years as a development scientist in high-throughput laboratories developing and improving biomarker and genomic assays for cancer diagnostics and genetic screening.

Share this article:

Related Content

Retuning the Heavens: Machine Learning and Ancient Astronomy

What can we learn about machine learning from ancient astronomy? When thinking about Machine Learning it is easy to be model-centric and get caught up…

Read More

Extracting Target Labels from Deep Learning Classification Models

In the blog post Configuring a Neural Network Output Layer we highlighted how to correctly set up an output layer for deep learning models. Here,…

Read More

True DX in the Pharma R&D Lab Defined by Enthought

Enthought’s team in Japan exhibited at the Pharma IT & Digital Health Expo 2022 life sciences conference in Tokyo, to meet with pharmaceutical industry leaders…

Read More

Exploring Python Objects

Introduction When we teach our foundational Python class, one of the things we do is make sure that our students know how to explore Python…

Read More

Choosing the Right Number of Clusters

Introduction When I first started my machine learning journey, K-means clustering was one of the first algorithms I was introduced to – and it is…

Read More

Prospecting for Data on the Web

Introduction At Enthought we teach a lot of scientists and engineers about using Python and the ecosystem of scientific Python packages for processing, analyzing, and…

Read More

Life Sciences Labs Optimize with New Digital Technologies and Upskilling

Labs are resetting the trajectory for drug development: reducing timelines from years to months; decreasing costs from billions to millions; and gaining an advantage by…

Read More

Configuring a Neural Network Output Layer

Introduction If you have used TensorFlow before, you know how easy it is to create a simple neural network model using the Keras API. Just…

Read More

No Zero Padding with strftime()

One of the best features of Python is that it is platform independent. You can write code on Linux, Windows, and MacOS and it works…

Read More

Got Data?

Introduction So, you have data and want to get started with machine learning. You’ve heard that machine learning will help you make sense of that…

Read More

Join Our Mailing List!

Sign up below to receive email updates including the latest news, insights, and case studies from our team.