ɬÀï·¬

The Impact of Research Software in Reproducible Neuroimaging Ìý

Presented by keynote speaker,ÌýÌýScientific Director of theÌý.Ìý

°ä´Ç²Ô³Ù±ð³æ³Ù:ÌýWe at the Faculty of Science'sÌýComputational and Data Systems Initiative (CDSI)Ìýhave been spending some time talking with various individuals and groups on campus with interests inÌýResearch Software Development. This includes faculty researchers, research software engineering staff on campus, and those involved with some of ɬÀï·¬'s software engineering education programs.ÌýÌýÌýÌý

What is beginning to emerge, to our eyes, is a sense ofÌýpotential synergyÌýamong these various groups. This mirrors, in turn, what seems to be an increasingly prevalent tendency worldwide on university campuses to try to create more of a critical mass around the research software development space, in contrast to the somewhat diaspora-like state that that community has typically had to date (e.g., witness the recent creation of theÌý).ÌýÌý

°Õ³ó±ðÌýResearch Software Development Networking SymposiumÌýis intended as anÌýinitial gatheringÌýof ɬÀï·¬'s professional research software development community, and related RSD/RSE-interested groups, to begin providing a sense ofÌýcross-campusÌýcommunity and to define mechanisms for continuing to develop and support that community going forward. Planned for 3hrs in length, the event will include aÌýkeynote, followed by a mix ofÌýlightning talksÌýfor group representatives to present the who/what/where/why/how of what they do andÌýstructured roundtable discussions. Food and drinks will be provided.Ìý

Please note that this event is primarily intended forÌýstaff in research software development roles on campus.

Keynote Presentation

Title: The impact of research software in reproducible neuroimagingÌý

Abstract: This talk first explores the impact of research software on neuroimaging analyses, focusing on the influence of hardware architectures, software dependencies, and software versions. It includes case studies on both traditional neuroimaging techniques and deep learning, with applications to MRI measures of Parkinson's disease. In the second part, we will highlight opportunities arising from software variability in scientific computing, such as the development of software tests, the use of ensemble models across different software environments, and potential for faster computations. Finally, we will discuss solutions to mitigate software variability, specifically through containerization frameworks and reproducible evaluation platforms.Ìý

Keynote speaker bio:ÌýÌýis the Scientific Director of theÌýÌýand a Senior Scientist at theÌýÌýin Toronto. He is also a Professor (status-only) in the Department of Psychiatry and an Associate Professor (status-only) in the Department of Computer Science at the University of Toronto (UoT). Additionally, Tristan holds an ongoing appointment as a Professor in the Department of Computer Science and Software Engineering at Concordia University in Montreal, where he held a Tier II Canada Research Chair in Big Data for Neuroinformatics. He has also served as a Visiting Scholar at ɬÀï·¬, a Research Scientist at the French National Centre for Scientific Research, and completed a Postdoctoral Fellowship at the University of Amsterdam. Tristan earned an Engineering degree from École Centrale de Lyon in France in 2004, followed by a Master of Science in Images and Systems from the same institution, and a PhD in Computer Science from the University of Nice Sophia-Antipolis in 2007. In 2022, he was elected to the Royal Society of Canada College of New Scholars.Ìý

Event Details

Event Date:ÌýThursday, April 10th 2025

Location:Ìý McIntyre Medical Building, room 325. See mapsÌý.

Schedule:Ìý

2:30 PM – 2:40 PM: Welcome Address & Event OverviewÌýÌý

2:40 PM – 3:50 PM: Keynote Speaker & Q&A –ÌýProf. Tristan GlatardÌý

3:50 PM – 4:00 PM: BreakÌýÌý

4:00 PM – 5:00 PM: Lightning TalksÌýÌý

5:00 PM – 5:20 PM: Lightning Brainstorm SessionÌý

5:20 PM – 5:30 PM: Conclusion, Q&AÌý

5:30 PM – 6:30 PM: Networking ReceptionÌý

Lightning Talk Proposals

If you are interested in presenting a 5-minute lightning talk, please provide us with a brief summary of your talk in the registration form below or send us your proposal by email at cdsi.science [at] mcgill.ca . Please note that only selected proposals will be contacted for further discussion.Ìý

Topical themes for Lightning Talks presentations include:

• CybersecurityÌý
• Research Data Management
• Policies and RegulationsÌý
• Research Data Storage Technologies (Hardware and Software)Ìý
• Machine Learning and Scientific ReproducibilityÌý
• Cross-domain mentorship, Training and CollaborationÌý