Different Devices, Unified Standards for Brain Cell Signals... Boosting Reliability of Animal-Free Testing [Reading Science]

KIT Develops Integrated Analysis Tool for MEA Data from Five Manufacturers
Source Code Released Free of Charge

A new tool has been developed to analyze the electrical signals of brain neurons, measured by devices from different manufacturers, using a unified standard. It is expected to improve the reproducibility and reliability of animal-free neurotoxicity tests utilizing brain organoids and other alternatives.

The Korea Institute of Toxicology (KIT) announced on September 9, 2026, that the research team led by Soonhyun Park, Director of the Center for Alternative Animal Research, along with teams from Soonchunhyang University, Sungkyunkwan University, and Pusan National University, has jointly developed the ‘CASCADE’ tool for integrated analysis of multi-electrode array (MEA) data.

CASCADE, which integrates and analyzes MEA data by manufacturer. It consolidates MEA data with different formats from each manufacturer into a single system to analyze neural activity and connectivity, neuronal avalanches, criticality, and other neural network information processing indicators. Provided by the research team

MEA is a device that simultaneously measures the electrical signals of cultured neurons or brain organoids across multiple electrodes. Although it is used in animal-free tests to assess the neurotoxicity of pharmaceuticals and chemicals, differences in data formats and analysis methods among manufacturers have made it difficult to directly compare results from different devices.

CASCADE processes MEA data from five manufacturers within a single analytical framework. By uploading the raw data, users can, with just one command line, analyze not only spikes and firing rates—the electrical signals and activity of neurons—but also criticality, which indicates the information processing state of the neural network, and neuronal avalanches, which describe the cascade-like spread of neuronal activity.

The research team validated CASCADE's performance using six real-world datasets from five manufacturers, including Axion BioSystems and MaxWell Biosystems. As a result, they were able to calculate and compare the criticality values of neural networks according to the same standard, regardless of differences in equipment or data format. The tool's spike detection performance for capturing neuronal electrical activity was comparable to that of existing manufacturer-dedicated software. Analysis took only about 5 to 20 seconds even on a general laptop, and even large datasets were processed with minimal memory usage.

Research team photo. (From left) Soonhyun Park, Head of the Animal Alternative Research Center at the Korea Institute of Toxicology (KIT), and Forbes Avila, Researcher. Provided by KIT

The team expects that this will enable laboratories using different equipment to compare experimental results on equal footing and improve the reproducibility of animal-free neurotoxicity assessments using cultured neurons and brain organoids. The source code and validation data have been released free of charge on GitHub, PyPI, and Zenodo.

Soonhyun Park, Director of the Center, stated, “In order to make scientific comparisons, neural signals must be analyzed by the same standard, regardless of the measuring device. We hope this contributes to advances in animal-free neurotoxicity testing and bio-intelligence research.”

The research results were published in the international journal ‘Bioinformatics’ on July 29, 2026.

This content was produced with the assistance of AI translation services.