Author ORCID Identifier

https://orcid.org/0000-0002-4127-9672

Document Type

Thesis

Date of Award

2024

Degree Name

Master of Science (MS)

Department

Biomedical Engineering

First Advisor

Etienne Z Gnimpieba

Abstract

The challenges faced while executing wet lab protocols encourage the development of automation systems to come alongside human scientists. Today’s cutting-edge experiments involve complex protocols with precise measurements usually performed manually. Even simpler biological protocols can be tedious and prone to error, as was seen during the COVID-19 pandemic and society’s demand for high-volume, rapid sample analysis. Moreover, reproducibility suffers when there is excessive variability and insufficient data. Here, we leveraged the Stanford Biodesign process to develop a modular lab automation system and image analysis workflow to address challenges like these. This flagship automation platform at The University of South Dakota’s Biomedical Engineering Department unlocks new possibilities. Since wet lab automation services are rare in the Midwest, our platform can ultimately bring more democratization and widespread regional adoption of bioscience automation. To evaluate our minimum viable product, we semi-automated basic eukaryotic and prokaryotic cell culturing, two widespread biology lab tasks susceptible to variability. By applying our system to these use cases, we learned of its—and our own—capabilities and areas for future improvement. We achieved protocol execution transparency via multi-angle video recording and workflow transparency with full protocol execution reports in Jupyter Notebook format. Additionally, projection toward FAIR Guiding Principles enhances data for transparency, quality control, and community reuse. Our system’s modularity minimizes the need for multiple inflexible systems, instead allowing for the removal of individual components as needed. Modularizing the software architecture will enable us to insert artificial intelligence and machine learning models, bringing more excellent value to our data. Soon, using open-source chatbots such as ChatGPT will assist in converting manual protocol steps to automated functions at a low cost. Deploying AI/ML opens doors to new opportunities for automated data analysis and iterative experimental design, which, when combined with automation hardware, is the genesis of self-driving labs and robot scientists.

Subject Categories

Artificial Intelligence and Robotics | Biomedical Engineering and Bioengineering

Keywords

AI/ML, automation, biodesign, cell culturing, image analysis, systems engineering

Number of Pages

114

Publisher

University of South Dakota

Available for download on Wednesday, September 03, 2025

Share

COinS