Ask for help about ADNI1 data

[ziru...@gmail.com]

Dears:

   I’m trying to download ADNI1 MRI data to conduct my research about AD.

   There is a place making me feel confused, i.e., why all “Visits” show “2” for "ADNI1:Baseline 3T”. For baseline, I thought all “Visits” should be “1”.

So is there anything wrong? Besides, what the relationship between “Visits” and their really month (i.e,0 month,3 month,6 month...)?

Yours sincerely,
Zi-Rui

[Danielle J Harvey]

Hi Zi-Rui,

 

Are you downloading images or downloading summaries derived from the images? The 3T scans in ADNI-1 were acquired at the baseline visit (there was also a screening visit, so that is probably why Visit is 2). 3T scans were only acquired on a subset of individuals, while everyone got 1.5T scans (the first one was at the screening visit). In ADNI-GO and ADNI-2, newly recruited participants were scanned at 3T. ADNI-3 is also using 3T.

 

If you are looking in the IDA (Image Data Archive) at image collections, I believe Visit is an internal number used by LONI. If you do a search in the IDA (rather than save images to a collection) and request a csv file, I believe that shows the visit (in terms of months). I brought this issue up to LONI a month or so ago, and I think they are trying to provide both their internal number as well as the month for both the IDA search and the IDA collections, so that is on their “to-do list”…

 

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 post to this group, send email to adni...@googlegroups.com.
Visit this group at https://groups.google.com/group/adni-data.
For more options, visit https://groups.google.com/d/optout.

[ZiRui W]

Dear Danielle:

    

    Very Thanks for your detailed instruction. 

    Yes, I am downloading both images and summaries from IDA Search - Data Collections.    

    Do you mean the first visit should be "ADNI1:Screening 1.5T" for these "ADNI1:Baseline 3T" ? 

    

      

    Besides, I want to know is what the different "MPR" and "MRP-R". 

     I notice there are "scale"  and "scale_2", which one I should use?   

    The last thing is I still not have the right to access ADNI2/GO (only ADNI1 are shown in my "Other  shared Collections").  

    Could you make me access to ADNI2/GO? 

    

Yours sincerely,

Zi-Rui

     

      

[Danielle J Harvey]

Hi Zi-Rui,

 

There have been collections set up for ADNI-1, but not yet for ADNI-GO or ADNI-2. The collections set up for ADNI-1 had to do with putting together sets of images that had no issues for the full sequence of visits of interest (see the following paper for information about the collections for ADNI-1, including some information about scale and scale_2)

 

Wyman BT, Harvey DJ, Crawford K, Bernstein MA, Carmichael O, Cole PE, Crane PK, DeCarli C, Fox NC, Gunter JL, Hill D, Killiany RJ, Pachai C, Schwarz AJ, Schuff N, Senjem ML, Suhy J, Thompson PM, Weiner M, Jack CR Jr.  Standardization of analysis sets for reporting results from ADNI MRI data.  Alzheimers and Dementia (2013);9:332-337. PMCID: PMC3891834

 

For ADNI-GO and ADNI-2 images, you’ll have to use the Advanced Search option in the IDA to acquire the images.

 

For ADNI1:Baseline 3T, the first 3T scan was done at visit=”bl” (baseline) which is visit #2. Most (if not all) of them will also have a 1.5T scan (in ADNI1:Baseline 1.5T) which will be at visit #1 (the screening visit). Since you are using those collections, you might find the spreadsheets downloadable from the Study Data page (where all other data files may be downloaded) useful, since those should include the visit codes in terms of months…

 

I believe MPR and MPR-R are for MPRAGE and MPRAGE repeat (if they needed to redo the scan).

 

Danielle

--

[ZiRui W]

Dear Danielle:

       Thank you very much.

       I got it!

       It's very kind of you !

Zi-Rui

--

To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+unsubscribe@googlegroups.com.

To post to this group, send email to adni...@googlegroups.com.
Visit this group at https://groups.google.com/group/adni-data.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "Alzheimer's Disease Neuroimaging Initiative (ADNI) Data" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/adni-data/-wbFK2SKkG8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to adni-data+unsubscribe@googlegroups.com.

[ZiRui W]

Dears:

     I'm sorry to bother you again.

     But I meet another question now.

     I feel confused about these different files for the same subject, i.e., Sag IR-SPGR, accelerated IR-SPGR, MPRAGE, MPRAGE SENSE2 and AGE GRAPPA2.           What's the difference between them?  Which one I should choose to conduct my research?    

     From their corresponding imaging protocol, I find only have small difference between these files, e.g., pixel sample.

Yours sincerely,

Zi-Rui

2018-01-31 14:56 GMT-05:00 Danielle J Harvey <djha...@ucdavis.edu>:

--

To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+unsubscribe@googlegroups.com.

To post to this group, send email to adni...@googlegroups.com.
Visit this group at https://groups.google.com/group/adni-data.
For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to a topic in the Google Groups "Alzheimer's Disease Neuroimaging Initiative (ADNI) Data" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/adni-data/-wbFK2SKkG8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to adni-data+unsubscribe@googlegroups.com.

[Danielle J Harvey]

In ADNI-2, we added an accelerated sequence (SENSE, SENSE2, GRAPPA) in addition to the non-accelerated sequences. IRSPGR or MPRAGE is specific to the vendor of the MRI (the "name" of essentially the same sequence). SENSE and SENSE2 are the accelerated version of the MPRAGE, while GRAPPA is the accelerated version of the IRSPGR.

You will need to decide whether you want to use the non-accelerated or the accelerated sequences (I wouln't use both or use accelerated for some individuals and non-accelerated for others). Once you decide which you want to use, then you can use both the IRSPGR and MPRAGE, since those are vendor specific - you shouldn't have both on a person, unless there was a scanner change that occurred sometime during follow-up.

Danielle

---

From: ZiRui W <ziru...@gmail.com>
Sent: Friday, February 2, 2018 10:02 AM
To: adni...@googlegroups.com; Danielle J Harvey

[ZiRui W]

Dear Danielle:

       Thank you very much! Always so kind. 

       There are some places making me feel confused

      

       When I use "Advanced Search" to search 3.0T1 original/pre-processed ADNIGO screening MRI, there are 334 original subjects, 146 pre-processed subjects. I find the 146 pre-processed subjects don't exist in the ADNI1 Data Collections. So I think maybe the 146 pre-processed subjects are the new subjects and the others are the same with the ADNI1. I am not sure whether I am right. 

      

        if the screening MRI is their first time, why their visit are "14"  for these ADNIGO screening MRI?

     

        Besides, I am sorry, where can I find their state (AD/EMCI or MCI or LMCI/NC) ? 

        

        Finally, according to your website, ADNIGO should contain 200 EMCI and 500 NC, MCI from ADNI1. But the number of my searching results is not matched. 

        Can you help me? 

       

Thank you very much.

Zi-Rui

[Danielle J Harvey]

Hi Zi-Rui,

 

The newly recruited participants in ADNI-GO were early-MCI (EMCI). All other participants were carry-overs from ADNI-1. One quick check is ADNI-1 participants have an RID < 2000 (the images have PTID which is generally in the format XXX_S_YYYY where YYYY is the RID); ADNIGO newly recruited participants have an RID in the 2000s and ADNI-2 newly recruited participants have an RID>=4000.

 

One other thing to keep in mind when looking at participants who carried over from ADNI-1 to ADNI-GO. Although a subset of individuals in ADNI-1 were scanned at both 1.5T and 3T, I believe that all ADNI-1 participants who rolled into ADNI-GO and ADNI-2 were continued to be scanned at 1.5T (not 3T), unless there was a major scanner change and the 1.5T scanner was no longer available.

 

I don’t know the mapping of visit codes to visit number, but I’m guessing that LONI had to keep all of the various visits separately identifiable, so even though for newly enrolled ADNI-GO participants, the screening MRI was the 1^st scan, the visit number is not 1.

 

The best place to find diagnosis information is to look on the Study Data page (where you can access all of the data files for ADNI).  When you log into the LONI ADNI website and click on Download, you have 3 options, one for Study Data, one for Image Collections and one for Genetics. I’m guessing you have gone to Image Collections and then used the Advanced Search…the Study Data page has a lot of .csv files that contain a variety of information on all participants in ADNI. In the first section,  you will find a file for Diagnostic Summary [ADNI2,GO,2,3]. This file contains diagnosis information (how it was recorded depends on which phase of ADNI, so ADNI-1 diagnosis information is in the variable DXCURREN, ADNI-GO & 2 have it in DXCHANGE and ADNI-3 has it in DIAGNOSIS). You can refer to the Data Dictionary for coding information (this file is found in the section for Study Info, Data & Databases on the Study Data page). In that same section (where the Data Dictionary is), there is also a file Key ADNI tables merged into one table (ADNIMERGE.csv) and corresponding dictionary file. This file contains the diagnosis information, possibly in a more easy to use format, since it collapses across the various variables used in the different phases of ADNI. The Roster file (in the Enrollment section of the Study Data page) has the link between PTID and RID (since RID is used in most of the files on the Study Data page).

 

As for the recruitment numbers, I believe the LONI website has what was originally planned in the proposal – the actual numbers of newly recruited participants and roll-over participants may be different.

 

I hope this helps!

Danielle

 

 

From: ZiRui W [mailto:ziru...@gmail.com]
Sent: Wednesday, February 7, 2018 10:05 AM
To: adni...@googlegroups.com; Danielle J Harvey <djha...@ucdavis.edu>
Subject: Re: [adni-data] Ask for help about ADNI1 data

 

Dear Danielle:

       Thank you very much! Always so kind. 

To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+...@googlegroups.com.

To post to this group, send email to adni...@googlegroups.com.
Visit this group at https://groups.google.com/group/adni-data.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "Alzheimer's Disease Neuroimaging Initiative (ADNI) Data" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/adni-data/-wbFK2SKkG8/unsubscribe.

To unsubscribe from this group and all its topics, send an email to adni-data+...@googlegroups.com.

[ZiRui W]

Dear Danielle:

       Great! Very detailed instruction.

       Thank you so much.

        

       I find a very few ADNI1-subjects have DXCONV (e.g., text line 42 from MCI to AD)  in baseline screening in the file of Diagnostic Summary [ADNI1,GO,2,3]. However, if these are their first screenings,  how possible ?  The similar case is a subject can change from AD to MCI in his base line (i.e., text line 388).  Besides, it seems impossible from AD to MCI or Normal.

       

       Thanks very much for your patience.

Yours sincerely,

Zi-Rui

To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+unsubscribe@googlegroups.com.

To post to this group, send email to adni...@googlegroups.com.
Visit this group at https://groups.google.com/group/adni-data.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "Alzheimer's Disease Neuroimaging Initiative (ADNI) Data" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/adni-data/-wbFK2SKkG8/unsubscribe.

To unsubscribe from this group and all its topics, send an email to adni-data+unsubscribe@googlegroups.com.

[Danielle J Harvey]

Hi Zi-Rui,

In ADNI, there was an initial diagnosis provided at the screening visit (when some information was collected to ensure that people met the inclusion criteria). This diagnosis was also used for randomizing individuals to arms of the study (1.5T MRI only, 1.5T+3T MRI, 1.5T MRI+FDG-PET). However, this diagnosis was based on limited information. At the baseline visit, a more complete assessment was made and another diagnosis was determined - the Clinical Core has recommended the use of the diagnosis at the baseline visit as the "initial" diagnosis -as you noted, there were a few changes in diagnosis between the screening and baseline visit, but these are to be based on more accurate information at the baseline visit, rather than an actual progression or reversion of the disease process.

So, for the baseline visit for ADNI-1, just use DXCURREN without worrying about whether there was a change in diagnosis from the screening visit.

Danielle

---

From: ZiRui W <ziru...@gmail.com>
Sent: Wednesday, February 7, 2018 2:25 PM
To: Danielle J Harvey; adni...@googlegroups.com

[ZiRui W]

Dear Danielle:

        Thank you so much.

        No language can describe my gratitude to you.

        Your contribution to humanity is too great.

Yours sincerely,

Zi-Rui

[ZiRui W]

Dear Danielle:

       Sorry to bother you again.

      

       There are some places I try to figure out.

       1.I use Advanced Search to download ADNI2 screening MRI. There are some different abbreviation I try to confirm, i.e, ADNI2 Screening MRI-New Pt, Initial Visit-Cont Pt, ADNI2 Screening-New Pt, ADNI2 Baseline-New Pt. I guess Pt means Patient. Initial Visit-Cont means from ADNI1/GO, New means new subjects and MRI- means the these images are MRI. It seems reasonable from my searching results.

       2.If I try to use "ADNI,ADNI Screening,ADNI Baseline,MRI,3.0T,Pre-processed" to search all baseline MRI data in ADNI. Only 130 subjects can be searched. But I can search 151  subjects from  ADNI1 3.0T screening and 146 new subjects from ADNIGO 3.0T screening.

       3. I know RID<2000 come from ADNI1, 2000<=RID<4000 are newly participants in ADNIGO and RID>=4000 are newly participants in ADNI2. Could you introduce some information about ADNI3?  

Thank you very much.

Yours Sincerely,

Zi-Rui  

        

[Danielle J Harvey]

HI Zi-Rui,

Sorry for the delay in responding - I was in meetings most of yesterday and then the email got buried...

You are correct in your interpretation of the different visits for ADNI2.

I have noticed in the IDA (where you can search for and download images), that some 3T scans are recorded as 2.9T (not sure why, but they are there). That might explain part of the discrepancy (for field strength, I put between 2.5 and 3.5 to capture all of them).

ADNI-3 newly recruited participants have RID>=6000.

Danielle

---

From: adni...@googlegroups.com <adni...@googlegroups.com> on behalf of ZiRui W <ziru...@gmail.com>
Sent: Thursday, February 8, 2018 10:25 AM

To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+...@googlegroups.com.

[ZiRui W]

Dear Danielle:

     It's fine. 

     Thank you very much. Always so helpful. 

      

      I try to use your method ( ADNI,ADNI Screening,ADNI Baseline,MRI,2.5T~3.5T,Pre-processed ) to search all baseline MRI data.  However, only ADNI1 (<2000) and ADNI3 (>6000) can be searched, i.e., there are 198 subjects (188 ADNI1, 10 ADNI3). According to my separate search, I can get 151 (the same number with your paper) subjects from ADNI1, 146 new subjects from ADNIGO, 840 new subjects from ADNI2. So it seems not to match.

    

    I hope you have a good rest. 

Yours sincerely,

Zi-Rui  

     

     

[ZiRui W]

Dear Danielle:

     It's fine. 

     Thank you very much. Always so helpful. 

      

      I try to use your method ( ADNI,ADNI Screening,ADNI Baseline,MRI,2.5T~3.5T,Pre-processed ) to search all baseline MRI data.  However, only ADNI1 (<2000) and ADNI3 (>6000) can be searched, i.e., there are 198 subjects (188 ADNI1, 10 ADNI3). According to my separate search, I can get 151 (the same number with your paper) subjects from ADNI1, 146 new subjects from ADNIGO, 840 new subjects from ADNI2. So it seems not to match.

    

    I hope you have a good rest. 

Yours sincerely,

Zi-Rui  

[Danielle J Harvey]

Hi Zi-Rui,

Are you using all of the possible visit names across the different studies? If you want to include ADNIGO and ADNI-2, you need to include ADNIGO Screening MRI, ADNI2 Screening-New Pt (in case some were classified not as an MRI visit), ADNI2 Screening MRI-New Pt, and ADNI2 Baseline-New Pt (probably not any, since all of the MRIs should have been done at the screening visit for ADNI-2).

When you do a separate search to get your 151 in ADNI-1, 146 in ADNIGO and 840 in ADNI2, what are you doing for the "separate search"?

Danielle

---

From: adni...@googlegroups.com <adni...@googlegroups.com> on behalf of ZiRui W <ziru...@gmail.com>

Sent: Friday, February 9, 2018 2:06 PM

To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+...@googlegroups.com.

[ZiRui W]

Dear Danielle:

        Thank you Danielle.

        Yes, it works if I include ADNIGO and ADNI-2.  

    

        Here we go. 

         I use Advanced Search "Phase: ADNI 1; Study/Visit: ADNI Screening, ADNI Baseline; IMAGE: MRI;  IMAGING PROTOCOL: 3.0 T1; IMAGE TYPES: Pre-processed" and the I can get 151 in ADNI-1.  

        I use Advanced Search "Phase: ADNI GO; Study/Visit: ADNIGO Screening MRI; IMAGE: MRI;  IMAGING PROTOCOL: 3.0 T1; IMAGE TYPES: Pre-processed" and the I can get 146 in ADNIGO.

         I use Advanced Search "Phase: ADNI 2; Study/Visit: ADNI2 Screening MRI-New Pt; IMAGE: MRI;  IMAGING PROTOCOL: 3.0 T1; IMAGE TYPES: Pre-processed" and the I can get 840 in ADNI-1.

Yours Sincerely,

Zi-Rui

[Danielle J Harvey]

Hi Zi-Rui,

So, just to confirm, you should get those same numbers (I did) if you select ADNI; Study/Visit: ADNI Screening, ADNI Baseline, ADNIGO Screening MRI, ADNI2 Screening MRI-New Pt; IMAGE: MRI;  IMAGING PROTOCOL: 3.0 T1; IMAGE TYPES: Pre-processed) and do the search all together, rather than do it separately by phases.

Danielle

---

From: ZiRui W <ziru...@gmail.com>
Sent: Monday, February 12, 2018 7:28 AM

[ZiRui W]

Dear Danielle:

        Interesting, if I search as your way rather than do it separately by phases, I can get 120 ADNI1, 146 ADNIGO and 840 ADNI2. Namely, the number of ADNI1 is not matched. The others are the same.

   

Yours Sincerely,

Zi-Rui