Memory and Knowledge (13 February 2008)
Sensitivity artefacts in MRI
- Combined spin echo and gradient echo images does not yield improvements in areas subject to sensitivity artefacts. Request for a tool to provide parameters for a sequence optimized for imaging a specific area.
Action: ChristianSchwarzbauer will talk to RikHenson about this.
Source control for software development
Action: MatthewBrett will give a Methods Group talk on Code Version Control on 14 February (see MethodsGroupMeeetings).
Imaging with patients
Advice on best practice in co-registration of brains with lesions: relative merits of manual masking of areas of lesion versus SPM5's default method.
Action: MatthewBrett will talk to KaralynPatterson about this.
Query about coregistration with MEG, and the use of MRI of lesioned brains to constrain MEG source solutions. Highlighing difficulty for SPM or MNE if there is a lesioned area on the cortical surface.
Action: None at present as MEG is outside the Methods Group's current range of expertise.
Methods Group's role in MEG development
Action: None at present as MEG is outside the Methods Group's current range of expertise. However noted that OlafHauk and OlegKorzyukov are active in expanding CBU expertise in this field, and we will look into case for expanding resources in this direction. Methods Group meetings will continue to provide a forum for sharing of experience and developing understanding.
Attention (19 February 2008)
ISSS sequence required for a number of studies (with Tristan and Matt)
- To present words in quiet;
To get multiple information on the time course of the hrf (especially with patients);
- Study on semantic ambiguity in subjects in vegetative state.
RhodriCusack asked about High Res fMRI.
- there was an issue over optimal voxel size to balance trade-off of thermal and physiological noise.
- cardiac gating
- 3 studies at CBU had use High Res
- this led to a discussion on sharing information to review what had been gained by using this protocol
- Zoom Imaging
- is available
- there had been a validation study with Kirsten
- suggested that we asked Geoff Aguirre about something ...
AdrianOwen asked what is the value of ASL (ArterialSpinLabelling).
- It can be use for short blocks (there is some means of finessing block effects?)
- Suggestion that combined EEG/fMRI be acquired for 'falling asleep' studies.
- there were technical problems with getting reliable EEG signal in the scanner; posssible the siutation would be better with sparse scanning.
- Possible benefits of 'spiral' sequences
- Use of sinusoidal gradient waveform for scanner acoustic noise reduction
- Real time acquisition
offered to meet AdrianOwen and others including CSL representatives
- wanted for as a kind of biofeedback medium for study on acute stroke patients switching tasks
JohnDuncan would invite Mark Stokes to come and talk about techniques that worked for him
RikHenson plans to invite someone from the Cambridge University Systems Group to talk
Language (26 February 2008)
Wanted very quiet acquisition sequences. CS suggested they could achieve this by use of different gradient waveforms e.g. sinusoidal, rather than implementation of the ISSS sequence.
Noise cancellation was raised as a potential help. This worked best below 500kHz. IHR were originally going to provide this.
Concern about contribution of acoustic noise to activation in frontal/temporal areas.
Discussion on headphones and their properties.
Possible use for event related parametric modulation.
Discussion on DWI as a short acquisition technique for identification of infarcted tissue.
PWI could be provided via ASL.
DanielWakeman is sourcing sequences for free from MGH for structural imaging. A wiki update relating to this would be sought.
Uses of sensitivity maps:
- To optimise imaging of a particular ROI
- To improve the power of group studies
Considering extending incorporation of sensitivity information into SPM-type analyses.
HIRES was mentioned.
MB is developing a DTI pre-processing tool.
Real time fMRI was discussed
- It is available now but not much used
- Data export is via new software from Siemens
- It is suitable for only a limited repertoire of designs
Might multi-echo EPI help to eliminate or at least reduce the impact of physiological noise?
How could the methods group contribute to MEG/EEG methods development? Discussion on MEG preprocessing and lack of a standard, and the absence of a single ideal analysis package.
Emotion (A few comments from PhilBarnard and AndyCalder by email; no meeting with the group to date)
- AC: One thing I'd discussed with members of E1 was being able to identify non-linear functions in SPM (e.g., a non-linear modulation of a particular contrast with personality dimension). I mentioned this briefly to Matthew and he thought it was possible.
- PB: Another longer term research issue discussed in the past with Ian was PCA's over time???