Article Figures & Data
Figures
- Fig 1.
Illustration of the automated CIV segmentation pipeline. This figure shows an infarct in the right hemisphere in a 6-day follow-up CT. A, The unprocessed NCCT. B, A seed point is positioned within the infarcted area by an observer. C, Determination of the midline. D, Segmentation of the ventricles, E, The final segmentation representing the CIV.
- Fig 2.
Example of incorrect automated segmentation due to hemorrhage and erroneous midline determination. A, NCCT scan of a patient with a midline shift. The blue line indicates the midline determined by the algorithm. The hemorrhage is indicated by the white arrow. B, The infarcted area delineated by observer 1. C, Incorrect outcome of automated segmentation. The hemorrhage is unrecognized as infarct, and the infarcted tissue is shifted over the original midline. This shift results in underestimation of the CIV.
- Fig 3.
Example of incorrect automated segmentation due to multiple infarcts. A, NCCT scan of a patient with multiple infarcts, 2 new infarcts (black arrows) and 1 old infarct (white arrow). B, The infarcted area delineated by observer 1. C, Incorrect outcome of automated segmentation. The old infarct and one of the new infarcts are unrecognized as infarcts. As a result, the automated segmentation underestimates the CIV.
Tables
- Table 1:
List of the number of patients per hospital, CT scanner used, and reconstruction kernel
Hospital No. of Patients CT Scanner Kernel 1 14 Sensation 64a H31s 2 2 Brilliance 40b UB 3 2 iCT 256 UB 4 1 Aquilion 64c FC23 5 1 Aquilion ONE FC26 6 3 iCT 256 EC 7 4 Sensation 16/ Somatom Definition Flash H31s 8 5 LightSpeed VCTd Soft 9 2 Brilliance 40 UB - Table 2:
Interobserver variability of manual CIV measurements and comparison of the manual and automated method
Correlationa (95% CI) Dice Coefficient (Mean and Range) Bland-Altman Limits of Agreement (mL) Relative Difference (SD) No. Automated vs manual 0.98 (0.97–0.99) 0.74 (0.42–0.90) (−38.0–39.1) 4% (± 20%) 34 Manual interobserver 0.98 (0.96–0.99) 0.84 (0.63–0.94) (−44.1–40.9) 11% (± 27%) 34 ↵a Pearson correlation; P < .01 for all coefficients.
In this issue
Jump to section
Related Articles
Cited By...
- Association of White Matter Lesions and Outcome After Endovascular Stroke Treatment
- Automatic segmentation of cerebral infarcts in follow-up computed tomography images with convolutional neural networks
- Comparison of three commonly used CT perfusion software packages in patients with acute ischemic stroke
- Hemorrhagic transformation is associated with poor functional outcome in patients with acute ischemic stroke due to a large vessel occlusion
- Association of follow-up infarct volume with functional outcome in acute ischemic stroke: a pooled analysis of seven randomized trials
- Impact of Ischemic Lesion Location on the mRS Score in Patients with Ischemic Stroke: A Voxel-Based Approach
- Incorporation of relative cerebral blood flow into CT perfusion maps reduces false at risk' penumbra
- Value of Quantitative Collateral Scoring on CT Angiography in Patients with Acute Ischemic Stroke
- Opercular Index Score: a CT angiography-based predictor of capillary robustness and neurological outcomes in the endovascular management of acute ischemic stroke
- Value of Thrombus CT Characteristics in Patients with Acute Ischemic Stroke
- Association of Computed Tomography Ischemic Lesion Location With Functional Outcome in Acute Large Vessel Occlusion Ischemic Stroke
- Baseline Blood Pressure Effect on the Benefit and Safety of Intra-Arterial Treatment in MR CLEAN (Multicenter Randomized Clinical Trial of Endovascular Treatment of Acute Ischemic Stroke in the Netherlands)
- Topographic distribution of cerebral infarct probability in patients with acute ischemic stroke: mapping of intra-arterial treatment effect
- Associations of Ischemic Lesion Volume With Functional Outcome in Patients With Acute Ischemic Stroke: 24-Hour Versus 1-Week Imaging
- Clot Burden Score on Baseline Computerized Tomographic Angiography and Intra-Arterial Treatment Effect in Acute Ischemic Stroke
- Influence of Device Choice on the Effect of Intra-Arterial Treatment for Acute Ischemic Stroke in MR CLEAN (Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands)
- Assessment of Collateral Status by Dynamic CT Angiography in Acute MCA Stroke: Timing of Acquisition and Relationship with Final Infarct Volume
- Thrombus Permeability Is Associated With Improved Functional Outcome and Recanalization in Patients With Ischemic Stroke
- Impact of Collateral Status Evaluated by Dynamic Computed Tomographic Angiography on Clinical Outcome in Patients With Ischemic Stroke