Trang Le
#math graduate. Postdoc fellow with Jason Moore.
Journal club
May 21st, 2019
1. extract image-related features
2. train model on normative dataset to predict age
3. obtain predicted ages on test set
4. compute
brain-predicted age difference
5. make inference: accelerated/
delayed aging
BAG (brain age gap) = predicted age - real age
brainPAD (brain-predicted age difference)
brainAGE (brain age gap estimate)
BAG ~ MS diagnosis
longitudinal analysis within MS
1,118 features
(360 ROIs: thickness, area, volume)
3,208 HC
xgboost
10-fold CV r = 0.91
235 HC
76 MS
Four types of Multiple Sclerosis (MS)
rate of aging
= \frac{\Delta \textrm{BAG}}{\Delta t}
Increased brain age gaps for all brain regions except temporal
Significant accelerated rate of brain aging compared to chronological aging
apparent accelerated aging of the brain may partly be explained by chronic inflammatory processes that drive neurodegeneration in MS
Questions
- circularity in the analysis of brain atrophy vs brain aging?
We estimated the annual global brain atrophy by comparing estimated total brain volume from the Freesurfer output (BrainSegVolNotVent) between time points.
Questions
- Estimation model accuracy: r = 0.91, what about MAE, MSE, etc.?
Discussion...
BrainAGE MS
By Trang Le
BrainAGE MS
Presentation on 2019-05-21 for the journal club, Kim Lab.
