NSRR Bootcamp Series: Sleep Analysis Open Source Toolboxes
Schedule:
| Presenter(s) | Workshop Title | Date | Time (EST) |
|---|---|---|---|
| Karine Lacourse, B.Ing. M.Sc.A. | The Snooz Toolbox: An Open-Access, Cross-Platform Solution for Sleep Data Analysis | January 23rd, 2025 | 1:00 PM |
| Luigi Fiorillo, Ph.D. | SLEEPYLAND: an open-source and easy-to-use toolbox to fairly evaluate automatic sleep staging algorithms on NSRR data | February 27th, 2025 | 1:00 PM |
| Gregory Hammad, Ph.D. | Powering Up Actigraphy Analyses with pyActigraphy | March 27th, 2025 | 1:00 PM |
| Joachim Behar, Ph.D.and Márton Áron Goda, Ph.D. | Advanced PPG Signal Analysis Using a Standardized Python Toolbox | April 29th, 2025 | 1:00 PM |
The event is free but space is limited and all attendees must apply:
Apply Here
Workshop Details
Powering up actigraphy analyses with pyActigraphy
March 27th, 2025 at 1pm EST
Apply Here
Overview: This workshop will guide users through the various actigraphy analysis steps; from data cleaning, batch computation of rest-activity rhythm variables to automatic rest episode detection.
Goals:
- To read and visualize actigraphy recordings
- To perform the necessary data cleaning steps
- To compute rest-activity rhythm variables
- To run multiple automatic rest detection algorithms and compute summary statistics
Resources:
- Online documentation: https://ghammad.github.io/pyActigraphy/index.html
- Git repository: https://github.com/ghammad/pyActigraphy
- Paper: Hammad G, Reyt M, Beliy N, Baillet M, Deantoni M, Lesoinne A, et al. (2021) pyActigraphy: Open-source python package for actigraphy data visualization and analysis. PLoS Comput Biol 17(10): e1009514. https://doi.org/10.1371/journal.pcbi.1009514
Prerequisites: Jupyter Notebook and python (3.6-3.10). Numpy and pandas. Additional packages will be installed automatically as dependencies of pyActigraphy.
Schedule
| Time (EST) | Topic |
|---|---|
| 1:00 PM - 1:10 PM | Quick overview of the current use of actigraphy in sleep and chronobiology research |
| 1:10 PM - 1:25 PM | Toolbox installation |
| 1:25 PM - 1:40 PM | Guided tour of the pyActigraphy software architecture |
| 1:40 PM - 2:00 PM | A picture is worth a thousand words: “How to read and visualize actigraphy data with pyActigraphy” |
| 2:00 PM - 2:10 PM | Tea break |
| 2:15 PM - 2:35 PM | Crap in, crap out: “How to clean actigraphy data with pyActigraphy” |
| 2:35 PM - 2:55 PM | Around the clock: “How to compute usual circadian rest-activity rhythm variables” |
| 2:55 PM - 3:10 PM | Sleep on it: “How to automatically detect and quantify sleep…hum hum.. rest in actigraphy recordings” |
Advanced PPG Signal Analysis Using a Standardized Python Toolbox
April 29th, 2025 at 1pm EST
Apply Here
Overview: pyPPG is a standard toolbox for real-time analysis of long-term finger PPG recordings. The toolbox extracts state-of-the-art PPG biomarkers (i.e., pulse wave features) from PPG signals. The algorithms implemented in the pyPPG toolbox have been validated on freely available PPG databases. Consequently, pyPPG offers a robust and comprehensive assessment of clinically relevant biomarkers from continuous PPG signals.
The following steps are implemented in the pyPPG toolbox:
- Loading a raw PPG signal
- Preprocessing
- Heartbeat detection
- Pulse wave segmentation
- Fiducial points identification
- Biomarker engineering
Goals:
- Overview of the pyPPG Toolbox
- Detection of standard fiducial points in PPG pulse waves using pyPPG
- Calculation and saving of pulse wave biomarkers from fiducial points
- Examples of applications in sleep physiological signals analysis
Resources: https://pyppg.readthedocs.io
Prerequisites: Python
Schedule
| Time (EST) | Topic |
|---|---|
| 1:00 PM - 1:15 PM | Introduction to the PhysioZoo resources for physiological time series analysis. Introduction to PPG background physiology and sensing techniques. |
| 1:15 PM - 1:30 PM | Toolbox installation |
| 1:30 PM - 1:45 PM | Basic command syntax |
| 1:45 PM - 2:00 PM | Hands-on examples using pyPPG: Example 1: Beat detection and pulse rate estimation, Example 2: Respiration rate estimation, Example 3: Feature engineering and machine learning. |
Speaker Bios
Karine Lacourse, B.Ing.M.Sc.A.
Algorithm Designer, Signal Analysis
Karine Lacourse, B.Ing., M.Sc.A., Algorithm Designer, holds extensive expertise in polysomnography (PSG) and signal processing. Her master’s research on source localization of cortical activity from magnetoencephalography (MEG) honed her skills in optimization and numerical modeling. Over the past nine years, she has developed and validated biometric algorithms in MATLAB and Python, including a sleep spindle detector designed to emulate human expert scoring. At CARSM (Center for Advanced Research in Sleep Medicine, ceams-carsm.ca), she contributed to creating a high-quality, open-source dataset of human-scored sleep spindles, advancing the validation of automated algorithms. Karine is also managing the Montreal Archive of Sleep Studies (MASS), an open-access and collaborative database of laboratory-based polysomnography (PSG) recordings. Currently, she is dedicated to the development, documentation, and training for the Snooz toolbox, an open-source platform for analyzing and interpreting PSG recordings.
Luigi Fiorillo, PhD
Postdoctoral Researcher, University of Applied Sciences and Arts of Southern Switzerland (SUPSI)
Dr. Luigi Fiorillo is a Postdoctoral researcher affiliated with the Biomedical Signal Processing Group at the University of Applied Sciences and Arts of Southern Switzerland. He completed his BSc. in Biomedical Engineering at the University of Naples Federico II in 2015, followed by an MSc. in Biomedical Engineering from the Polytechnic University of Turin in 2018. In 2022, he earned his PhD in Computer Science from the University of Bern. During his doctoral studies, he was based at the Computer Vision Group laboratory under the mentorship of Paolo Favaro. His primary responsibility was the development and implementation of deep learning algorithms for projects in the field of sleep medicine. He also spent a year at the Martin Monti laboratory at UCLA in California, contributing to develop a real-time automatic sleep scoring and sleep patterns monitoring algorithm during LIFU stimuli tuning. His research interests primarily revolve around the application of Machine Learning and Deep Learning in the domains of sleep medicine and neuroscience. Preliminary work "U-Sleep’s resilience to AASM guidelines" (NSRR Blog, npj digital medicine article )
Grégory Hammad, PhD
GIGA - CRC In vivo Imaging, University of Liège, Liège, Belgium, Faculty of Medicine, Technical University of Munich, Munich, Germany
As an elementary particle physicist who moved to neuroscience a few years ago, I am interested in studying the link between aging and circadian, but also ultradian, rhythms in physiological signals. Link to the online documentation of our actigraphy analysis tool
Joachim A. Behar, PhD
Assistant Professor, Biomedical Engineering, Technion-Israel Institute of Technology
Dr. Behar founded the Artificial Intelligence in Medicine Laboratory (AIMLab.) at the Technion Faculty of Biomedical Engineering in March 2019. His work involves research of medical artificial intelligence to benefit patient care, and includes development of machine learning algorithms for analyzing large medical unstructured, i.e., physiological time series and medical images and datasets, with an emphasis on early diagnosis in the medical disciplines of cardiology, sleep medicine and ophthalmology. He has made significant contributions to the discovery of diagnostic biomarkers in cardiology and sleep medicine. The lab has already conducted large-scale experiments on extensive databases, encompassing millions of participants and thousands of hours of continuous physiological recordings. Dr. Behar published authored over 50 journal papers in top life science and engineering journals. Since 2016 he serves as editor for the journal Physiological Measurement and has been on the program committee of the Computing in Cardiology international conference since 2013. Dr. Behar is twice winner of the MIT-PhysioNet/Computing in Cardiology competition. In 2022, he initiated the Technion-Rambam Initiative in Medical AI (TERA), a collaborative effort between the Technion and Rambam Health Care Campus, utilizing large medical datasets and state-of-the-art AI advances to combat human diseases. Lab website
Márton Áron Goda, PhD
Faculty of Biomedical Engineering, Technion-IIT
Dr. Márton Áron Goda completed his PhD in 2021 and founded the Acoustic Imaging and Medical Signals Laboratory at Pázmány Péter Catholic University. He also began co-supervising a PhD student he had previously mentored during their master’s degree. In 2022, he accepted a postdoctoral position at the Technion Faculty of Biomedical Engineering in Israel, joining the Artificial Intelligence in Medicine Laboratory. He conducted his postdoctoral research there from August 2022 to August 2024 to establish long-term collaboration between the two institutions.
At the Technion, Dr. Goda led several successful research projects. He developed a high-performance PPG detector, improving its speed by 57 times using Python. This led to a collaboration with the University of Cambridge, resulting in an open-source Python toolbox called pyPPG, which was later integrated into PhyzioZoo software. Since its release in 2024, pyPPG has been downloaded over 4,100 times and cited in 11 publications. It is now widely used in various research projects.
In another project with Rambam Hospital, he studied the link between fetal breathing movements and lung maturity estimation. During this time, he and his PhD student also created an open-source toolbox for PCG signal analysis.
Dr. Goda is now an assistant professor at Pázmány Péter Catholic University, leading his laboratory and building collaborations with other research groups and medical institutions."
Past Sessions:
2025 NSRR Bootcamp Series
Snooz Toolbox, an open-access, cross-platform solution for sleep data analysis
SLEEPYLAND: an open-source and easy-to-use toolbox to fairly evaluate automatic sleep staging algorithms on NSRR data
NSRR Winter Webinar Series: Sleep Data Analysis Showcase
PhysioZoo: Utilizing an Open Digital Resource of Physiological Biomarkers in Sleep Medicine
Probing Complex Physiologic Signals During Sleep: Applications to Assessing Neuroautonomic Function
SleePyLand: A Python Library to Analyse the Large Amount of NSRR Sleep Data via Deep Learning Algorithms
Luna: A Toolset for Largescale Sleep Signal Analysis
Analysis of Rest-Activity Patterns with the NSRR: The Case of the MESA Actigraphy Dataset
Creating Sleep EEG Fingerprints for Biomarker Development Using the DYNAM-O Toolbox
Analysing the Cyclic Alternating Pattern of NREM Sleep
2023 Winter Webinar Series: Harnessing the Power of Animal Data for Sleep and Circadian Science
Session 1- The NSRR as a logical place to deposit animal-based sleep/wake recordings
Monday, December 4, 2023 at 3-3:30 PM ET
Session 2- Metabolic interventions for sleep in aging
Monday, December 4, 2023 at 3:30-4 PM ET
Session 3- Cellular and Animal Measurement of Circadian Clocks
Monday, December 4, 2023 at 4-4:30 PM ET
Session 4- Sleep and memory in the rat brain
Monday, December 4, 2023 at 4:30-5 PM ET
Session 1 - NSRR: Unlocking the Power of Sleep Data
Date: Wednesday, September 7, 2022 at 2 PM ET
Session 2 - Data Wrangling and Management: How to Optimize the Value of your Data
Date: Wednesday, October 12, 2022 at 2 PM ET
Session 3 - The New NIH Data Sharing Policy: How the NSRR Can Help
Date: Thursday, November 17, 2022 at 3 PM ET (rescheduled from 11/3/2022)