API?

[Michael Romano]

Hello!

I have been going through the metadata for a while now, and have found very specific dates/scans in which I am interested. It would be much easier if I were able to query the individual scans programmatically i.e. via a python script. Is there any mechanism available to do this right now? My current pipeline is just to download everything I can, then find the ones that I need, and then go from there. Thank you!

Best,

Mike

[Danielle J Harvey]

Hi Mike,

 

Apologies for the delay in responding. At this time, I believe the Advanced Search is the only way to identify scans, unless you just download everything and then do your script on those files. I’m not sure which cases are of particular interest to you to provide a specific recommendation to an alternative strategy. I can tell you that LONI is working on an entirely new data query system, which will enable users to create their own datasets, identify the images of interest, etc. and then download them. I beta tested the system late last year and it has a lot of great functionality. I’m not sure when the new system will be rolled out, but once it is, I think it will make everything a lot easier.

 

Danielle

--
You received this message because you are subscribed to the Google Groups "Alzheimer's Disease Neuroimaging Initiative (ADNI) Data" group.
To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/adni-data/63dc7bc2-58f8-487b-bbd8-68123274d5c8n%40googlegroups.com.

**CONFIDENTIALITY NOTICE** This e-mail communication and any attachments are for the sole use of the intended recipient and may contain information that is confidential and privileged under state and federal privacy laws. If you received this e-mail in error, be aware that any unauthorized use, disclosure, copying, or distribution is strictly prohibited. If you received this e-mail in error, please contact the sender immediately and destroy/delete all copies of this message.

[Michael Romano]

Ah, thank you so much!

So the data we are looking at is very specific. For example, we are looking at dates where patients converted to AD, and then identifying the MRI scans that were conducted 1 year prior. Then, we are looking at baseline MRIs for patients who did not convert to AD.

I am currently finding MRIs using the '*' wildcard. For example, *FSPGR*accel*. Are there other wildcards that are available for querying? For example, is there a way to do something like *MP*RAGE* | *FSPGR* (find all either MPRAGE or FSPGR)?

[Danielle J Harvey]

Hi Mike,

 

Apologies for the delay – been dealing with some big grant deadlines! The query system is definitely not currently set up in a way to make such a search all that straightforward. Using the wildcard ‘*’ is definitely a way to go about it – you just want to make sure you consider the different types of T1 descriptors (for example, we also have IR-SPGR). Are you interested in using the raw images or ones that have already undergone some corrections by the MRI Core? Are you looking across all phases of ADNI or restricting to a certain phase?

 

You can restrict to T1 images (if you select the imaging protocol from the list on the left of the screen to be included as part of the search options), which might help, if you aren’t already doing that.

To view this discussion on the web visit https://groups.google.com/d/msgid/adni-data/75cf665d-70a6-473c-8d29-2778696f77c9n%40googlegroups.com.

[Michael Romano]

Hi Danielle,

No worries at all!

Ok, just wanted to make sure there wasn't a trick I was missing. We are using all phases, raw MRI scans and raw PET images, so unfortunately are looking for tau images via *tau* and MRIs by clicking the T1 box and doing *accel*spgr*, *accel*mp*rage*, *grappa2*, and *sense2*. If the desired subject isn't found, I do some deeper digging, and if the subject still has nothing, I use the full mris instead of the accelerated mris. Inefficient, but it's all I've figured out so far. I've been successful querying specific subjects using comma separation though. I originally used partially pre-processed PET images, but when I went to register them with the raw MRIs the world-coordinates got kind of messed up, so now I'm just processing tau from scratch.

Thanks again!

Mike