Related Experiment Video
Updated: Nov 23, 2025
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.1K
RippleNet: a Recurrent Neural Network for Sharp Wave Ripple (SPW-R) Detection
Espen Hagen1,2, Anna R Chambers3, Gaute T Einevoll4,5
1Faculty of Science and Technology, Norwegian University of Life Sciences, Ås, Norway. espenhgn@gmail.com.
Neuroinformatics
|January 4, 2021
Summary
We developed RippleNet, an AI tool using deep learning to accurately detect hippocampal sharp wave ripples (SPW-R), crucial for memory. This method automates detection from brain signals, improving research efficiency.
Area of Science:
- Neuroscience
- Artificial Intelligence
- Computational Biology
Background:
- Hippocampal sharp wave ripples (SPW-R) are vital biomarkers for memory consolidation and decision-making.
- Accurate SPW-R detection is essential for understanding brain function in health and disease.
Purpose of the Study:
- To introduce a novel, self-improving AI-based method for accurate SPW-R detection.
- To contrast the AI approach with conventional, manual methods for SPW-R identification.
Main Methods:
- Utilized deep Recurrent Neural Networks (RNN) with Long Short-Term Memory (LSTM) layers for feature learning from raw local field potential (LFP) data.
- Employed supervised learning on hand-curated datasets of SPW-R events from the hippocampus (CA1 region).
- Implemented a non-causal (bidirectional) variant of the algorithm, named RippleNet, for enhanced accuracy.
Main Results:
- The AI method accurately detects SPW-R events by learning features directly from LFP data.
- The bidirectional RippleNet variant demonstrated superior accuracy compared to its unidirectional counterpart.
- The algorithm outputs time-varying probabilities of SPW-R events, enabling precise identification via thresholding.
Conclusions:
- RippleNet offers a highly precise and automated solution for SPW-R detection, surpassing traditional methods.
- The open-source availability and ease of integration facilitate its adoption in neuroscience research workflows.
- This AI-driven approach enhances the study of neural mechanisms underlying cognition and behavior.
