[Paper Review] An open-source implementation of a closed-loop electrocorticographic Brain-Computer Interface using Micromed, FieldTrip, and PsychoPy

Bob Van Dyck, Arne Van Den Kerchove+1 more|arXiv (Cornell University)|Feb 10, 2026

EEG and Brain-Computer Interfaces0 citations

TL;DR

The paper presents an open-source, modular Python-based stack for closed-loop ECoG BCIs that integrates Micromed acquisition, FieldTrip real-time processing, and PsychoPy-based experiment control, including three libraries (psychopylib, pymarkerlib, pyfieldtriplib) and runnable examples.

ABSTRACT

We present an open-source implementation of a closed-loop Brain-Computer Interface (BCI) system based on electrocorticographic (ECoG) recordings. Our setup integrates FieldTrip for interfacing with a Micromed acquisition system and PsychoPy for implementing experiments. We open-source three custom Python libraries (psychopylib, pymarkerlib, and pyfieldtriplib) each covering different aspects of a closed-loop BCI interface: designing interactive experiments, sending event information, and real-time signal processing. Our modules facilitate the design and operation of a transparent BCI system, promoting customization and flexibility in BCI research, and lowering the barrier for researchers to translate advances in ECoG decoding into BCI applications.

Motivation & Objective

• Provide a detailed technical description of a closed-loop ECoG-based BCI deployment in a clinical Micromed setting.
• Introduce three open-source Python libraries (psychopylib, pymarkerlib, pyfieldtriplib) to enable flexible, modular BCI development.
• Demonstrate integration of these tools with compact runnable examples that cover training, calibration, and online use.
• Promote transparency, customization, and lower barriers for translating ECoG decoding advances into BCI applications.

Proposed method

• Modular Python-based workflow combining data acquisition, real-time processing, and a user application.
• Three libraries: psychopylib for structured PsychoPy experiment design, pymarkerlib for sending precise event markers, pyfieldtriplib for thread-based real-time signal processing pipelines.
• Real-time data streaming via FieldTrip buffer with a MATLAB proxy, using TCP/IP/WLAN for data transfer and synchronization.
• Epoching, preprocessing, and modeling implemented in real-time (RtEpoch, RtFunction) with a multithreaded pipeline to avoid blocking acquisition.
• demonstrations include an open/close movement classification use case with a motor-imagery paradigm and a closed-loop feedback mechanism.
• Latency considerations decomposed into ADC, processing, and output stages; strategies to bound or reduce latency and to set a suitable decision rate.

Experimental results

Research questions

• RQ1 How can a modular, open-source Python stack support closed-loop ECoG BCI experiments in a clinical setting using Micromed hardware?
• RQ2 What are the practical design decisions and software tools that enable precise stimulus timing, synchronization, and real-time processing in ECoG BCIs?
• RQ3 Can three lightweight libraries (psychopylib, pymarkerlib, pyfieldtriplib) successfully implement end-to-end BCI pipelines from data acquisition to closed-loop feedback?
• RQ4 What are the latency and decision-rate trade-offs in such a modular system, and how can they be managed in asynchronous vs synchronous use?

Key findings

• The authors provide three open-source Python libraries (psychopylib, pymarkerlib, pyfieldtriplib) that enable experiment design, event synchronization, and real-time signal processing for closed-loop ECoG BCIs.
• A detailed system description shows integration of Micromed acquisition, FieldTrip buffer, and PsychoPy-driven experiments in a clinical context, including a FieldTrip proxy and MATLAB-MEX interface.
• They demonstrate runnable examples including a closed-loop movement classification use case and a motor-imagery EEG use case, with end-to-end latency analyses.
• The Appendix use cases illustrate training, calibration, and asynchronous BCI use with explicit epoch-based processing and feedback mechanisms.
• The work emphasizes modularity and minimalism, with future work aiming to remove MATLAB dependency by porting FieldTrip Micromed proxy to Python.