Integrating clinical and imaging features to predict recurrence of cerebrovascular events —— A machine learning study
mengting wei1, jinhao lv2, liuxian wang2, senhao zhang2, dongshan han2, xinrui wang2, and xin lou2
1Chinese PLA General Hospital, BeiJing, China, 2Chinese PLA General Hospital, beijing, China
55 patients were enrolled in this study. After
feature selection, 11 variables were included in the study, and the first three
variables with greater contribution were HCR, mRS on admission and diabetes history. FVH
score and collateral circulation grade had no significant contribution to
recurrent stroke. Then it was divided into model A and model B according to
whether the HCR is included or not. The results show that RandomForest and
NaiveBayes are the optimal algorithms to identify patients with recurrent
cerebrovascular events within one year through machine learning. In addition,
there were significant differences between model A and model B.
Man,
73 years old, Paroxysmal weakness of both lower limbs for 5 months.
Figure
A ,PWI raw image; Figure B , MRA showed right middle cerebral artery occlusion;
Figure C ,Tmax 4s was 238.04ml, Tmax 6s was
113.60ml, Tmax 8s was 29.66ml, Tmax 10s was 4.42ml ( HCR=47.72% ; HIR = 3.89%).
Comparison between
models