In a significant breakthrough for the field of personalised medicine, scientists from the Singapore-MIT Alliance for Research and Technology (SMART) have developed a novel technique that rapidly detects microbial contamination in cell cultures using ultraviolet (UV) spectroscopy and machine learning.
The method, unveiled by SMART’s Critical Analytics for Manufacturing Personalized-Medicine (CAMP) interdisciplinary research group, promises to shorten sterility testing times from days to just 30 minutes. This could accelerate the delivery of critical cell therapy products (CTPs), especially for patients with life-threatening conditions.
The study, conducted in collaboration with the Massachusetts Institute of Technology (MIT), ASTAR Skin Research Labs (ASRL), and the National University of Singapore (NUS), was published in Scientific Reports under the title “Machine learning aided UV absorbance spectroscopy for microbial contamination in cell therapy products”.
By analysing the unique UV “fingerprints” of cell culture fluids, the team trained machine learning models to distinguish contaminated samples with high accuracy. This process requires only a small volume of sample and eliminates the need for cell extraction, incubation periods, or growth enrichment mediums.
“This rapid, label-free method is designed to be a preliminary step in the CTP manufacturing process as a form of continuous safety testing, which allows users to detect contamination early and implement timely corrective actions,” said Shruthi Pandi Chelvam, senior research engineer at SMART CAMP and lead author of the study.
Current standard sterility tests can take up to 14 days, a delay that can be life-threatening for critically ill patients in urgent need of treatment. Even more advanced rapid microbiological methods (RMMs), while quicker, still take about seven days and require skilled operators and complex procedures.
The new approach developed by SMART not only reduces the time and labour involved but also supports the automation of sampling and analysis. This could help streamline the entire manufacturing process for cell therapies and reduce the risk of human error.
“Traditionally, cell therapy manufacturing is labour-intensive and subject to operator variability,” said Professor Rajeev Ram, Principal Investigator at SMART CAMP and corresponding author of the paper. “By introducing automation and machine learning, we hope to streamline cell therapy manufacturing and reduce the risk of contamination.”
The technique is also cost-effective, requiring no specialised or expensive equipment, and has the potential for wider applications beyond medicine. Researchers say future work will focus on expanding the system’s ability to detect a broader range of contaminants and validating its performance across different cell types and manufacturing environments.
Additionally, the method could be adopted in sectors like food and beverage manufacturing, where microbial quality control is critical to consumer safety.
The research is supported by the National Research Foundation (NRF) Singapore through its Campus for Research Excellence and Technological Enterprise (CREATE) programme. 