HFO-artifact

HFO-Classifier Model Card

<div align="center"> <img src="https://github.com/roychowdhuryresearch/pyHFO/blob/main/src/ui/images/icon1.png?raw=true" width="12%" alt="PyHFO" /> </div>

High-Frequency Oscillation (HFO) Classification Models

This repository hosts a suite of machine learning models designed to classify high-frequency oscillations (HFOs) in neural signals. HFOs are critical biomarkers often studied in neurological research, particularly in the context of epilepsy and brain function. The models in this repository are tailored for different classification tasks to help researchers and clinicians analyze HFOs more effectively.

Models Included

1. Artifact Detection Model: Identifies and filters out artifacts from HFO signals.

2. spkHFO Detection Model: Detects spike-associated HFOs (spkHFOs), a specific subtype of HFOs associated with epileptic spikes.

3. eHFO Detection Model: Identifies epileptic HFOs (eHFOs), another subtype strongly linked to epileptogenic regions of the brain.

Key Features

• Comprehensive Classification Pipeline: Covers artifact removal, spkHFO detection, and eHFO detection to streamline HFO analysis workflows.
• State-of-the-Art Models: Built using advanced deep learning techniques to ensure high accuracy and robustness.
• Easy-to-Use API: Models can be loaded directly using Hugging Face's transformers library for seamless integration into research pipelines.