Related Experiment Video
Updated: May 5, 2026
07:22
Optical Detection of E. coli Bacteria by Mesoporous Silicon Biosensors
Published on: November 20, 2013
16.9K
Machine Learning-Assisted Liquid Crystal Optical Sensor Array Using Cysteine-Functionalized Silver Nanotriangles for
Maryam Mousavizadegan1, Morteza Hosseini1, Mohammad Mohammadimasoudi2
1Nanobiosensors Lab, Department of Life Science Engineering, Faculty of New Sciences and Technologies, University of Tehran, Tehran 1439817435, Iran.
ACS Applied Materials & Interfaces
|December 12, 2024
Summary
A new liquid crystal (LC)-based optical sensor rapidly identifies common bacteria in food and water. This cost-effective platform uses machine learning for accurate bacterial differentiation, offering a simple solution for public health.
Area of Science:
- Biotechnology
- Nanotechnology
- Analytical Chemistry
Background:
- Rapid bacterial identification in food and water is crucial for public health.
- Existing methods can be time-consuming or require specialized equipment.
- A need exists for simple, cost-effective, and accurate bacterial detection systems.
Purpose of the Study:
- To develop a single-probe liquid crystal (LC)-based optical sensing platform for differentiating common bacterial strains.
- To utilize cysteine-functionalized silver nanotriangles as signal enhancers for improved detection.
- To apply image analysis and machine learning (ML) for pattern recognition and accurate identification.
Main Methods:
- A liquid crystal (LC) optical sensing platform was designed using cysteine-functionalized silver nanotriangles.
- Unique optical patterns were generated upon interaction with bacterial samples.
- Image analysis coupled with machine learning algorithms, particularly Support Vector Machines (SVM), was employed for pattern recognition.
Main Results:
- The platform successfully differentiated five common bacterial strains (Bacillus cereus, Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, and S. typhimurium).
- Support Vector Machines achieved a high accuracy of 98.89% in bacterial differentiation.
- The sensor demonstrated a linear range of 10-10^6 CFU mL^-1 and detection limits below 10 CFU mL^-1.
- High prediction accuracies were observed in real-world samples: water (95.83%), juice (97.92%), and milk (89.58%).
Conclusions:
- The developed LC-based optical sensing platform offers a simple, cost-efficient, and highly accurate method for bacterial identification.
- The integration of silver nanotriangles and machine learning significantly enhances detection capabilities.
- This technology holds promise for routine monitoring of bacterial contamination in food and water safety applications.
