Automatic segmentation of hemispheric CSF on MRI using deep learning: Quantifying cerebral edema following large hemispheric infarction
Junzhao Cui,
Jingyi Yang,
Ye Wang,
Meixin Ma,
Ning Zhang,
Rui Wang,
Biyi Zhou,
Chaoyue Meng,
Peng Yang,
Jianing Yang,
Lei Xu,
Guojun Tan,
Lidou Liu,
Junli Zhen,
Li Guo,
Xiaoyun Liu
Affiliations
Junzhao Cui
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Jingyi Yang
Department of Data Center, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Ye Wang
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Meixin Ma
University of California, Berkeley College of Letters and Science, US
Ning Zhang
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Rui Wang
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Biyi Zhou
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Chaoyue Meng
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Peng Yang
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Jianing Yang
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Lei Xu
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Guojun Tan
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Lidou Liu
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Junli Zhen
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Li Guo
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China
Xiaoyun Liu
Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, China; Department of Neurology, The First Hospital of Hebei Medical University, Shijiazhuang, China; Corresponding author. Department of Neurology, The First Hospital of Hebei Medical University, Shijiazhuang Donggang Road, Yuhua District, Shijiazhuang City, Hebei Province, China.
Background and objective: Cerebral edema (CED) is a serious complication of acute ischemic stroke (AIS), especially in patients with large hemispheric infarction (LHI). Herein, a deep learning-based approach is implemented to extract CSF from T2-Weighted Imaging (T2WI) and evaluate the relationship between quantified cerebrospinal fluid and outcomes. Methods: Patients with acute LHI who underwent magnetic resonance imaging (MRI) were included. We used a deep learning algorithm to segment the CSF from T2WI. The hemispheric CSF ratio was calculated to evaluate its relationship with the degree of brain edema and prognosis in patients with LHI. Results: For the 93 included patients, the left and right cerebrospinal fluid regions were automatically extracted with a mean Dice similarity coefficient of 0.830. Receiver operating characteristic analysis indicated that hemispheric CSF ratio was an accurate marker for qualitative severe cerebral edema (area under receiver-operating-characteristic curve 0.867 [95% CI, 0.781–0.929]). Multivariate logistic regression analysis of functional prognosis showed that previous stroke (OR = 5.229, 95% CI 1.013–26.984), ASPECT≤6 (OR = 13.208, 95% CI 1.136–153.540) and low hemispheric CSF ratio (OR = 0.966, 95% CI 0.937–0.997) were significantly associated with higher chances for unfavorable functional outcome in patients with LHI. Conclusions: Automated assessment of CSF volume provides an objective biomarker of cerebral edema that can be leveraged to quantify the degree of cerebral edema and confirm its predictive effect on outcomes after LHI.
