Related Experiment Video
Updated: Jul 23, 2025
11:07
High-Throughput Metabolic Profiling for Model Refinements of Microalgae
Published on: December 4, 2021
3.8K
An artificial intelligence approach for identification of microalgae cultures
P Otálora1, J L Guzmán1, F G Acién2
1University of Almería, CIESOL, ceiA3, Department of Informatics, 04120 Almería, Spain.
New Biotechnology
|July 19, 2023
Summary
A new artificial neural network model accurately characterizes microalgae cultures using FlowCam imaging. This deep learning approach offers a fast, reliable method for monitoring microalgae in large-scale production.
Area of Science:
- Biotechnology
- Marine Biology
- Artificial Intelligence
Background:
- Microalgae culture characterization is crucial for biomass quality control.
- Existing methods can be time-consuming and lack speed for large-scale applications.
- Developing efficient tools is essential for optimizing microalgae production.
Purpose of the Study:
- To develop a simple, fast, and accurate model for microalgae culture characterization.
- To utilize artificial neural networks for distinguishing between different microalgae genera.
- To enhance the reliability of microalgae biomass quality assessment.
Main Methods:
- Development of an artificial neural network model.
- Data acquisition using FlowCam for cell image capture.
- Training the model with multiple species across 6 microalgae genera.
- Implementation of a classification threshold to improve accuracy.
Main Results:
- The model achieved up to 97.27% accuracy in classifying microalgae cultures.
- Successfully distinguished between 6 different genera of microalgae.
- The classification threshold effectively discarded unwanted objects, boosting performance.
Conclusions:
- Deep learning models are effective tools for microalgae culture characterization.
- The developed model provides accurate and rapid monitoring for large-scale facilities.
- This method supports quality assurance in microalgae biomass production across diverse genera.
