Understanding on Baseline data columns in ADNIMERGE dataset

[Debopriya Ghosh]

Hi Experts,

I have downloaded ADNIMERGE.csv file from ADNI database and planned to use it for my research on Early prediction of alzheimer's disease using Machine Learning techniques. I have gone through the ADNIMERGE dataset and understand that it is having data for different visit codes. But I want to do analysis on Base Line data. To filter that I am using VISCODE = bl. But in the spread  sheet there are few columns which are named with and without bl. Example, EXAMDATE_bl and EXAMDATE, CDRSB_bl and CDRSB, ADAS11_bl and ADAS11, ADAS13_bl and ADAS13, MMSE_bl and MMSE, DX_bl and DX. When I filtered by VISCODE = 'bl', for same RID, few of the column of _bl and without _bl are having different entries. For analysis Baseline data, which such columns ("with 'bl' ", or "without 'bl' ") should we refer. Please suggest. Also the base line phase comes after the screening phase . Is my understanding correct? Please suggest!

[Naomi Saito]

Hello,

I downloaded ADNIMERGE file this morning to check if we have different values for those variables you mentioned. 

I used VISCODE == "bl" rows, and found NO differences in EXAMDATE and EXAMDATE_bl, CDRSB and CDRSB_bl, ADAS11 and ADAS11_bl, ADAS13 and ADAS13_bl, MMSE and MMSE_bl.  (I took summary of differences (i.e. datedif = EXAMDATE - EXAMDATE_bl,  cdrdif = CDRSB - CDRSB_bl), and they are all zero). 

Could you download the file, and check again?

For diagnosis (DX and DX_bl), DX has 3 categories (CN, MCI, Dementia), and DX_bl has 5 categories (CM, SMC, EMCI, LMCI, AD:  CN and SMC belong to CN group, and EMCI and LMCI belong to MCI group). 

I see that N=9  (N=7:ADNI1, N=1:ADNIGO, N=1:ADNi2) have different diagnosis between DX and DX_bl.  I checked DXSUM data (DXCURREN, DXCONV, DXREV for ADNI1, DXCHANGE for ADNIGO2), and found that DX variable has correct baseline diagnosis  (They have conversion/reversion coding at baseline in DXSUM data). 

For ADNI3, DX and DX_bl have same diagnosis. 

Yes, screening is done before baseline.  You can also check "SCHEDULES" tab from the link below.

http://adni.loni.usc.edu/adni-3/

Naomi

ADNI | ADNI 3 DOCUMENTS. The full list of ADNI documents can be found on the Documents page.New documents for ADNI 3 can be seen by clicking the links below. Study data and additional password-protected documents can be found in the Image and Data Archive, with instructions for access on the Access Data page.. Standardized MRI Data Sets adni.loni.usc.edu

---

From: adni...@googlegroups.com <adni...@googlegroups.com> on behalf of Debopriya Ghosh <debo...@gmail.com>
Sent: Wednesday, October 6, 2021 12:47 AM
To: Alzheimer's Disease Neuroimaging Initiative (ADNI) Data <adni...@googlegroups.com>
Subject: [adni-data] Understanding on Baseline data columns in ADNIMERGE dataset

 

Hi Experts,

I have downloaded ADNIMERGE.csv file from ADNI database and planned to use it for my research on Early prediction of alzheimer's disease using Machine Learning techniques. I have gone through the ADNIMERGE dataset and understand that it is having data for different visit codes. But I want to do analysis on Base Line data. To filter that I am using VISCODE = bl. But in the spread  sheet there are few columns which are named with and without bl. Example, EXAMDATE_bl and EXAMDATE, CDRSB_bl and CDRSB, ADAS11_bl and ADAS11, ADAS13_bl and ADAS13, MMSE_bl and MMSE, DX_bl and DX. When I filtered by VISCODE = 'bl', for same RID, few of the column of _bl and without _bl are having different entries. For analysis Baseline data, which such columns ("with 'bl' ", or "without 'bl' ") should we refer. Please suggest. Also the base line phase comes after the screening phase . Is my understanding correct? Please suggest!

--
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/99de406a-5b4f-447c-9b36-dd0bf28889cdn%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.

[Debopriya Ghosh]

Thanks nhsaito for the information.

So, if I want use this ADNIMERGE csv file for AD stage prediction(CN,MCI,AD etc.) using Machine Learning Techniques, should I use DX column as target or DX_bl column as target variable?

Thanks 

[Naomi Saito]

I would use DX.

Naomi

---

From: adni...@googlegroups.com <adni...@googlegroups.com> on behalf of Debopriya Ghosh <debo...@gmail.com>

Sent: Saturday, October 9, 2021 9:55 AM

To: Alzheimer's Disease Neuroimaging Initiative (ADNI) Data <adni...@googlegroups.com>

Subject: Re: [adni-data] Understanding on Baseline data columns in ADNIMERGE dataset

 

To view this discussion on the web visit https://groups.google.com/d/msgid/adni-data/517fa095-ef06-42f3-8901-4f6ad746b3cen%40googlegroups.com.

[Warda Saeed]

I think that depends on your research question. How fine grained you want it. If it’s mere MCI/AD comparison then follow what Naomi suggested. 

To view this discussion on the web visit https://groups.google.com/d/msgid/adni-data/BY3PR08MB712361DE7A553801BC3D99C0B9B59%40BY3PR08MB7123.namprd08.prod.outlook.com.

[Debopriya Ghosh]

Thanks  Wardas and Naomi!

My Research Question is identification of early stages of AD, in that can I should use DX_bl correct, because this column is having more granular stages/sub groups than DX...correct?

Thanks