- Thursday, January 7th 2016 at 16:00 - 17:00 UK (Other timezones)
- General participation info | Participate online | + Phone in United States: +1 (571) 317-3129 Australia (toll-free): 1 800 193 385 Australia: +61 2 8355 1020 Austria (toll-free): 0 800 202148 Belgium (toll-free): 0 800 78884 Canada (toll-free): 1 888 455 1389 Denmark (toll-free): 8090 1924 France (toll-free): 0 805 541 047 Germany (toll-free): 0 800 184 4222 Israel (toll-free): 1 809 454 830 Italy (toll-free): 800 793887 Netherlands (toll-free): 0 800 020 0182 New Zealand (toll-free): 0 800 47 0011 Norway (toll-free): 800 69 046 Poland (toll-free): 00 800 1213979 Portugal (toll-free): 800 819 575 Spain (toll-free): 800 900 582 Switzerland (toll-free): 0 800 740 393 United Kingdom (toll-free): 0 800 169 0432 United States (toll-free): 1 877 309 2073 Access Code: 731-636-357 Audio PIN: Shown after joining the meeting Meeting ID: 731-636-357
The aim of the TCPW is to bring the nascent field of computational psychiatry (CP) together in an innovative way to help have a greater impact in academic psychiatry. For the January 2016 meeting, we would therefore like to suggest one specific step in this direction and set up the Computational Psychiatry Multisite Data Project (CPMDP).
CP consists of several independent labs that use their own bespoke tasks and/or approaches to modelling. This is nice, but brings some problems. For CP to have an impact, people have to be able to use the tools, and we need to reduce the barriers for those starting out. It also makes it more difficult to directly compare results from different laboratories and to select among the various modelling / fitting approaches. Most critically, the reliability and normative properties of measures across labs remain unclear, a problem which is of great importance given the clinical aspirations of the field. The CPMDP aims to address these issues.
The idea is to:
- Jointly (at TCPW meetings) decide on a task of broad interest to us as a community, and on a metric to evaluate models of it
- Provide code to run the task in all labs
- Have all the labs collect data with the task (and obtain some minimal subject data – NIH CDE)
- Share all data
- Have each lab analyse all the data with their own favourite modelling approach
- Jointly evaluate analyses on the performance metrics and settle on one standard modelling approach.
What would this give us?
- A benchmarked task the robustness of which has been characterised across sites
- A documented critical assessment of modelling and fitting approaches, which can also serve as tutorial for those entering the field
- Example data on the selected task including
- consistency of model parameters between sites
- a database of raw data which can be used in future analyses
If you’re interested
Then please prepare brief statements on:
- Your favourite task you’d like to run through this pipeline; why we should run it; and how a model of it should be evaluated. Please note that for the first iteration of this, we are looking for a SIMPLE task!
- A brief overview over tricky ethics requirements you may face so that we can harmonise these across labs
- How to best share the financial / workload between labs, and how to apportion (what type of) reward.