A bioinspired microbial taste chip with artificial intelligence-enabled high selectivity and ultra-short response...

Area of Science:

• Environmental Science
• Biosensor Technology
• Artificial Intelligence

Background:

• Global water pollution poses a significant threat to human health.
• Microbial taste chips offer potential for real-time water monitoring but face challenges in sensor selectivity and response time.
• Existing gene-editing techniques have not fully resolved selectivity issues, with response times often exceeding 3 minutes.

Purpose of the Study:

• To develop a bioinspired wireless microfluidic microbial taste chip with enhanced selectivity and reduced response time for water pollution monitoring.
• To address the critical challenge of sensor selectivity in microbial taste chip technology.
• To enable accurate and real-time detection of specific heavy metal ions in water.

Main Methods:

• Development of a bioinspired wireless microfluidic microbial taste chip.
• Application of artificial intelligence (AI), specifically gated recurrent unit (GRU)-based deep learning algorithms.
• Analysis of temporal output current patterns generated by the chip in response to different pollutants.

Main Results:

• Achieved 98.9% classification accuracy for Cu²⁺, Pb²⁺, and Cr⁶⁺.
• Demonstrated a shortest response time of 48 seconds, which is 3.75 times faster than previous benchmarks.
• The chip enables real-time sensing with high accuracy and linearity for targeted heavy metal ions.

Conclusions:

• The developed AI-enabled microbial taste chip offers a breakthrough solution for selective and rapid water pollution monitoring.
• The chip's small footprint, wireless connectivity, and high performance make it suitable for real-time quantitative heavy metal ion detection.
• This technology can significantly contribute to global efforts in combating water pollution and ensuring water safety.