Cedalion Workshop: Python for fNIRS & DOT Analysis

Cedalion is a new Python-based toolbox designed for advanced multimodal fNIRS analysis and Diffuse Optical Tomography (DOT) that can be collaboratively expanded by the community. The toolbox is developed by the IBS Lab (BIFOLD) in close collaboration with the BOAS Lab at the BU Neurophotonics Center — © IBS Lab

Cedalion Workshop

Python-based analysis of fNIRS and DOT in the everyday world with multimodal methods and machine learning

Join us for a 1.5-day workshop on Cedalion, a Python-based toolbox designed for advanced multimodal fNIRS and DOT analysis that can be collaboratively expanded by the community. Following Cedalion’s main release in 2026, the workshop will familiarize researchers with the foundations of the toolbox. Cedalion integrates and expands core features of HOMER3 and AtlasViewer's processing and image reconstruction while tapping into the rich Python ecosystem of modern tools for data analysis and machine learning.

Date: August 24. & 25., 2026
Venue: Hybrid (Virtual & onsite). Onsite: BIFOLD / Technische Universität Berlin, Franklin Str. 28/29, 7th Floor, Room 701.
Registration: via Congeno-Form

At the end of the workshop, participants will have learned:

Foundations of fNIRS acquisition and experiments
How to use Cedalion for state-of-the-art channel- and image space analysis, preprocessing, regression-based GLM analysis, head models, photon simulation and image reconstruction, photogrammetric probe co-registration, and how to employ Cedalion’s support of machine learning methods for multimodal analysis and classification of fNIRS signals.

Proposed course Schedule

Time zone in CEST. May be subject to change

Time	Aug 24th	Aug 25th
	Day 1: Foundations & Channel Space Analysis	Day 2: Image Space & Machine Learning
08:00	Arrival & Set Up	Arrival & Set Up
08:30	Welcome & Workshop Orientation	Head Models, Photon Simulation & DOT Image Reconstruction
08:45	Introduction to fNIRS: Theory and Hands-On	Head Models, Photon Simulation & DOT Image Reconstruction
10:10	Coffee Break	Coffee break
10:30	Installation, Data I/O, Core Data Structures	Advanced multimodal Analysis & Machine Learning
11:15	Basic Preprocessing	Advanced multimodal Analysis & Machine Learning
12:30	Lunch	END
13:30	Quality Assessment & Motion Correction
14:30	General Linear Model (GLM)
15:30	Coffee Break
15:45	Processing Pipelines & HomerGUI
17:15	END

Registration

Two participation modes are offered: in-person and virtual.

Cedalion Workshop Registration

Register here

In-person attendance

Limited to 35 seats; Seats are assigned first-come, first-served.
A seat is confirmed only after:
1. registering for an in-person slot, and
2. paying a €50 fully refundable commitment deposit. Alternatively, seats can be confirmed with proof of booked travel documents.
The €50 deposit is refunded in full on the first day of the workshop upon on-site check-in.
Registration without payment does not reserve a seat.
After the 35 seats are confirmed
1. In-person registration closes and no further deposits are accepted.
2. Additional registrants are placed on a waiting list without deposit.
3. If a confirmed in-person participant cancels, a waiting-list participant may be offered the seat on short notice, otherwise they can attend virtually.

Virtual attendance

Open to all participants.
No capacity limit, no deposit required.
No on-site support and Q&A

Administrative matters

Target attendees
Participants are expected to have basic knowledge in Python. Foundational or advanced knowledge of fNIRS and fNIRS analysis is not necessary but can be a plus.

Hard/Software Requirements for attendees
Participants should bring their own laptops with a running Python installation on their machine. Cedalion is distributed under a permissive MIT license, and no paid licenses are required to take part. Participants are encouraged to install the Cedalion toolbox before the workshop (instructions will be provided 2 weeks prior). Participants are welcome to bring their own fNIRS dataset for experimenting with the toolbox (SNIRF data format required); example datasets are also provided.

Recommended Reading