Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Cellular and Molecular Bioengineering

Cellular and Molecular Bioengineering - Poster Session D

Poster I6 - Rapid Bacterial Detection in Wastewater via Colorimetric Analysis of Dye-Integrated Copolymer with Machine Learning-Based Quantification

Friday, October 25, 2024
3:30 PM - 4:30 PM EST
Location: Exhibit Hall E, F & G

Introduction:  Waterborne pathogens contribute significantly to global health issues, with the CDC estimating 7.15 million illnesses, 118,000 hospitalizations, and 6,630 deaths annually, costing approximately $3.3 billion in healthcare. E. coli, a common indicator of water contamination, includes harmless and pathogenic strains like E. coli O157, which can cause severe symptoms such as bloody diarrhea and stomach cramps. To tackle this issue, we use a novel antibacterial copolymer, PCTAD, integrated with a dye to detect elevated bacterial concentrations in wastewater rapidly. Our approach combines this with a machine-learning model that predicts bacterial concentrations based on light absorbance data from a microplate reader. This model outputs bacterial concentrations in CFU/ml and correlates them with specific color intensities, enabling precise and timely monitoring of bacterial concentration. This integrated method offers an effective solution for rapid bacteria detection and provides a standardized colorimetric response corresponding to bacterial concentrations.

Materials and

Methods:  We hypothesize that as PCTAD interacts with bacteria in wastewater, the amount of free PCTAD available to react with potassium iodide (KI) dye decreases. Consequently, the solution's color intensity will be inversely proportional to the bacterial concentration, with more intense color indicating lower bacterial levels.
Bacterial concentrations in wastewater were first quantified using agar plate assays with a ten-fold dilution series. The efficacy of PCTAD against wastewater microbes was assessed via the streak plate method, using sterile deionized (DI) water as a negative control and an empty agar plate as a positive control. To remove dust particles from our wastewater sample and ensure bacteria were not left behind in the residue, we used gravity filtration with a filter paper diameter of 150 mm. The filtrate was then used to validate the hypothesis.
We validated our hypothesis with a 96-well plate assay. The first row served as a control with DI water, while subsequent rows contained wastewater samples. Twelve serial dilutions of PCTAD were performed, maintaining consistent bacterial concentrations in each experimental well. A 30-minute wait time was allowed before adding the KI dye to ensure sufficient reaction between PCTAD and bacteria.
The 96-well plate was then placed in a Tecan Infinite M200 plate reader to measure light absorbance at 600nm. The absorbance data was used to train and validate a linear regression machine learning model. The model's accuracy was further validated and tested with unseen test data.

Results, Conclusions, and Discussions: Our experiment demonstrated that PCTAD effectively eliminates bacterial contamination in wastewater, with initial bacterial concentrations measured at 10^6 CFU/ml, as evidenced by no colony formation in the control groups and significant bacterial presence in the experimental group. Additionally, the experimental group in the 96-well plate assay exhibited an earlier color change in the dilution sequence compared to the control group.
Plate reader data revealed decreased light absorbance with decreasing PCTAD dilutions, correlating with higher bacterial concentrations. Despite this, the experimental group showed slightly higher absorbance than the control, suggesting residual bacterial presence. Furthermore, the linear regression machine-learning model demonstrated high predictive accuracy with an R-squared value of 0.9635 and a mean square error of 0.0125. This model effectively correlated bacterial concentrations with colorimetric results, enhancing the visualization of bacterial levels.

The color change difference between PCTAD dilutions in the control and experimental groups is likely due to the consumption of PCTAD by the bacteria in the wastewater. Since less free PCTAD was available for the KI dye to react with, a clear color change was observed earlier in the PCTAD dilution sequence for those wells with bacteria than for those without bacteria which supports our hypothesis.

In conclusion, we have demonstrated that the novel antimicrobial material PCTAD can be used for effective bacterial detection in wastewater samples. Furthermore, a machine-learning model has been developed to quantify bacterial concentration using plate reader data, associating that with specific colors to provide a rapid and visual detection method. This approach shows promise for rapid bacterial detection and quantification in sectors such as food service, natural resource management, and pharmaceutical manufacturing.

Acknowledgements (Optional):  We extend our sincere gratitude to the Department of Biomedical Engineering at the University of South Dakota for their support. Special thanks to Pawan Sigdel, a graduate student at the South Dakota School of Mines and Technology (SDM&T), for providing the wastewater samples crucial to this study. We also wish to acknowledge the invaluable contributions of the staff and personnel in Dr. Gnimpieba’s lab for their assistance throughout the project.