AI facilitated fluoro-electrochemical phytoplankton classification.

Area of Science:

• Marine Biology
• Biotechnology
• Environmental Science

Background:

• Marine phytoplankton are crucial for global carbon cycling and oxygen production.
• Accurate identification of phytoplankton is vital for monitoring ocean health and climate change.
• Current methods struggle with the diversity and scale of phytoplankton species.

Purpose of the Study:

• To develop a novel method for distinguishing phytoplankton taxonomies.
• To leverage artificial intelligence for automated analysis of phytoplankton characteristics.
• To enhance the capability for autonomous ocean monitoring.

Main Methods:

• Utilized fluoro-electrochemical microscopy to analyze chlorophyll-a fluorescence quenching in seawater.
• Developed a neural network for analyzing fluorescence transients.
• Tested classification accuracy on 29 phytoplankton strains.

Main Results:

• Chlorophyll-a fluorescence quenching rates proved characteristic of phytoplankton species.
• The neural network achieved >95% accuracy in classifying phytoplankton to taxonomic orders.
• The combined method offers high granularity and flexibility.

Conclusions:

• Fluoro-electrochemical microscopy combined with AI provides a state-of-the-art solution for phytoplankton classification.
• This approach is adaptable for autonomous ocean monitoring systems.
• The method significantly advances our ability to study marine phytoplankton and their ecological roles.