For now, what I want to do with this post is remove from the article certain sections and subsections in which I only have notes and ideas written down for before I finish the article. There are quite a few things I have to polish in SCCs (to call it something quickly), and here I am going to list the stuff I must not forget before publication:

1-Limitations and Further Research

I have found a series of interesting articles that point towards new avenues of research and also the opposite: what appear to be insurmountable issues that will have to be navigated. Some of these references should be discussed in this section, even if only to put into context what has been done to date and also to understand how the SCC methodology fits within present and future research and its potential impact.

These limitations and proposals for future research include:

PET lacks standard measures and does not have a consensual treshold for diagnosis
Given the connection between glucose metabolism and mitochondria, and the fact that mitochondrial DNA is maternally inherited in humans, it is here argued that altered bioenergetics may be an upstream event in those with a maternal history of late-onset AD link
Concerns about stablished standardizations
Costs: As of August 2008, Cancer Care Ontario reports that the current average incremental cost to perform a PET scan in the province is 1,000{1,200$ per scan. This includes the cost of the radiopharmaceutical and a stipend for the physician reading the scan.[75 Wiki].
There are new promising venues for early diagnosis of AD by means of other diagnostic methods such as blood analysis detecting levels of phosphorylated Tau proteins. Thijssen EH et al. “Diagnostic value of plasma phosphorylated tau181 in Alzheimer’s disease and frontotemporal lobar degeneration”. Nature Medicine.
Also new treatments such as: PI3K activation prevents AB42-induced synapse loss and favors insoluble amyloid deposit formation.

2-Notes for Article refinement:

Add cross-refs between sections.
Append Code for different sections or at least refer to the different folders in GitHub.
Include MCI group in our comparisons (?).
I should also cite the nifti packages.
Create a colour palette that is the one used in Biomedicine in general.
Add self-citation as a self-sustained positive feedback in MDD re/AD
Keep an active voice and use the present tense
Remove excesive ”it is”, ”the’s”, ”that’s”.
Place adjectives before nouns
Target regions include amygdala. Future research.
Brodmann area 20 == temporal cortex? Regions highlighted in figure 1 are precise enough?
Complete GitHub repository
Be careful with the citation of the GitHub repository; I have not found a standard.
Add details about the PET data (specifications) and their processing (for example, mean average scaling and other parameters used in preprocessing)

3-Comparative Table:

What comes next is to make a comparative table with the turquoise plots of all the comparisons I have. The question is how to represent it visually in an attractive and succinct way. In the article by Wang et al. (2019), what they do is divide the image into 3 columns and 5 rows. The 3 columns are because, by using MCI, they have three comparisons to make: ADvsCN, ADvsMCI, CNvsMCI. But in my case we only have AD and CN, so it does not make sense there. Then the 5 rows are for 1 whole group and 4 subsets: ‘female’, ‘male’, <75, >75.

What I actually have to show in an image right now would be:

AD vs CN:
- for the complete group
- for women
- for men
- for under 75
- for over 75

Right now, what I am considering is the optimal way to present the results graphically, which is really something that concerns me quite a lot, since I think that a good part of the study’s power lies in how the subject is visualised. I think that in duplicate (in two strategic Zs) it would gain quite a lot, and it would gain even more if I also used the MCI subjects.

For now, what I am going to do today is present the changes by sex and by age.

< Back to home