Background and Purpose. The gray-to-white matter ratio (GWR) on brain computed tomography (CT) is associated with neurological outcomes after cardiac arrest (CA); however, the prognostic value of GWR in CA patients has yet to be confirmed. Therefore, we conducted a meta-analysis of related studies to investigate the prognostic value of GWR on brain CT for neurological outcomes after CA. Materials and Methods. The PubMed, ScienceDirect, Web of Science, and China National Knowledge Infrastructure databases were searched for all relevant articles published before March 31, 2020, without any language restrictions. The pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated with a random-effects model using Stata 14.0 software. Result. A total of 24 eligible studies with 2812 CA patients were recruited in the meta-analysis. The pooled result showed that decreased GWR was correlated with poor neurological outcomes after CA (
OR
=
11.28
, 95% CI: 6.29–20.21, and
P
<
0.001
) with moderate heterogeneity (
I
2
=
71.5
%
,
P
<
0.001
). The pooled sensitivity and specificity were 0.58 (95% CI: 0.47–0.68) and 0.95 (95% CI: 0.87–0.98), respectively. The area under the curve (AUC) of GWR was 0.84 (95% CI: 0.80–0.87). Compared with GWR (cerebrum) and GWR (average), GWR using the basal ganglion level of brain CT had the highest AUC of 0.87 (0.84–0.90). Subgroup analysis indicated that heterogeneity may be derived from the time of CT measurement, preset specificity, targeted temperature management, or proportion of cardiac etiology. Sensitivity analysis indicated that the result was stable, and Deeks’ plot showed no possible publication bias (
P
=
0
.64
). Conclusion. Current research suggests that GWR, especially using the basal ganglion level of brain CT, is a useful parameter for determining neurological outcomes after CA.
Deep Learning-enabled Detection of Acute Ischemic Stroke using Brain Computed Tomography Images
