FIOLA: an accelerated pipeline for fluorescence imaging online analysis

Area of Science:

• Neuroscience
• Biophysics
• Computational Biology

Background:

• Optical microscopy, including calcium and voltage imaging, allows for rapid monitoring of neuronal activity in large populations.
• A bottleneck exists in analyzing this imaging data quickly, hindering real-time applications and rapid experimental iteration.
• Current online algorithms struggle to keep pace with high-throughput imaging speeds and indicator dynamics.

Purpose of the Study:

• To develop a novel framework for rapid online analysis of fluorescence imaging data.
• To overcome the limitations of existing algorithms in extracting neuronal activity from calcium and voltage imaging movies.
• To enable real-time closed-loop experiments and accelerate the cycle of experiment-analysis-theory.

Main Methods:

• Developed FIOLA (Fluorescence Imaging Online analysis), a software framework optimized for parallel processing on GPUs and CPUs.
• Implemented algorithms designed for high-speed extraction of neuronal activity from fluorescence imaging data.
• Validated FIOLA's performance using both simulated and real-world calcium and voltage imaging datasets.

Main Results:

• FIOLA achieves neuronal activity extraction speeds one order of magnitude faster than current state-of-the-art methods.
• Demonstrated reliable and scalable performance across diverse imaging datasets.
• Presented a practical online experimental scenario to guide parameter selection and illustrate trade-offs.

Conclusions:

• FIOLA significantly advances the online analysis of fluorescence imaging data, addressing a critical bottleneck in neuroscience.
• The framework supports real-time applications and facilitates faster scientific discovery in high-throughput neuroimaging.
• FIOLA offers a powerful, efficient, and scalable solution for extracting neuronal activity from optical microscopy data.