Premium
P3‐347: INDIVIDUAL EVALUATION SYSTEM FOR WHITE MATTER HYPERINTENSITY RECOGNITION USING DEEP CONVOLUTIONAL NEURAL NETWORK
Author(s) -
San Lee Jin,
Lee Kyung Mi,
Kim Eui Jong,
Rhee Hak Young,
Park Key-Chung,
Oh Jang-Hoon,
Lee Jae Ho,
Lee Hyun Sub,
Kim Hyug-Gi
Publication year - 2018
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2018.06.1708
Subject(s) - hyperintensity , convolutional neural network , artificial intelligence , computer science , pattern recognition (psychology) , medicine , magnetic resonance imaging , radiology
linear mixed effect model.Results:Among 65 CAApatients, 43 (66.2 %) showed A}$^[fish 0,mfnt]>b PET positivity. Typical cases of A}$^[fish 0,mfnt]>b (-) CAA are shown in Figure 1. A}$^[fish 0,mfnt]>b (+) CAA patients have a higher number of lobar MBs (24.1 6 29.8 vs. 7.0 6 8.1, p 1⁄4 0.010) and a higher frequency of CSS (34.9 vs. 9.1%, p1⁄40.025), while A}$^[fish 0,mfnt]>b (-) CAA patients have a higher number of lacunes (0.8 6 1.7 vs. 2.5 6 3.9, p 1⁄4 0.017) and a higher frequency of severe WMH (20.9 vs. 45.5%, p1⁄40.040). (Table 1) Finally, A}$^[fish 0,mfnt]>b positivity was associated with faster decline in multiple cognitive domains including language (p < 0.001), visuospatial function (p < 0.001), verbal memory (p < 0.001), and general cognition (p < 0.001) in a linear mixed effects model. (Figure 1) Conclusions:Our findings suggested that A}$^[fish 0,mfnt]>b positivity in CAA patients might represent advanced CAA burdens, which affect worse cognitive trajectory.