Article Figures & Data
Figures
- Fig 1.
Automatic extraction of CSF space. The figures in the top row show the MIP images on the T2-weighted 3D-SPACE sequence in the representative iNPH case. Light green indicates the subarachnoid space segmented automatically at a threshold intensity of >700 on the SYNAPSE 3D workstation. The other figures show the 3D volume-rendering reconstruction images of the subarachnoid space on the second line, total CSF on the third line, and ventricles on the last line. The left, middle, and right column figures show axial, coronal, and sagittal dimensional views, respectively.
- Fig 2.
Division of the subarachnoid space into the upper and lower parts. The subarachnoid space was divided into the upper and lower parts in a horizontal section on the anterior/posterior commissure plane at the level of the junction point of the vein of Galen and the straight sinus. The left and right figures show coronal and sagittal views. The Sylvian fissure is typically included in the lower part of the subarachnoid space. The upper-to-lower subarachnoid space ratio was defined as the upper part to the lower part of the subarachnoid space.
- Fig 3.
Division of the subarachnoid space into the 4 parts. The subarachnoid space was divided into the following 4 parts: frontal convexity (yellow), parietal convexity (magenta), Sylvian fissure and basal cistern (sky blue), and posterior fossa (light green) in the 3D segmentation. The left, middle, right upper 3D volume-rendering reconstruction images show the axial, coronal, and sagittal dimensional views, respectively. Light green in the left lower axial MIP image indicates the segmented region of the parietal convexity of the subarachnoid space, and that in the middle and right lower MIP images indicates axial and coronal views of the Sylvian fissure and basal cistern.
- Fig 4.
Three-directional linear indicators for evaluating the size of bilateral ventricles. The figures are the 3-directional MPR reconstruction images of the T2-weighted 3D SPACE. On the basis of the anterior/posterior commissure line, x-, y-, and z-axes for spatial coordinates of head position were defined. The green lines show the maximum length of 3-axial directions of the bilateral ventricles. The yellow lines show the maximum 3-axial length of the frontal horn of the bilateral ventricles. The red lines show the maximum intracranial 3-axial width. In addition to the original Evans Index, the y- and z-Evans Indices were defined as the maximum length from the foramen of Monro to the anterior and superior extremities of the frontal horns (yellow lines)/the maximum intracranial y- and z-axial length (red lines), respectively. The x-, y-, and z-Maximum Indices were defined as the maximum width of the bilateral ventricles (green lines)/the maximum intracranial width on the each of the 3 dimensions (red lines).
- Fig 5.
Receiver operating characteristic curves for discriminating tap-positive from the tap-negative group. The ROC graphs show specificity on the x-axis and sensitivity on the y-axis. The left graph shows the ROC curve of the parietal convexity of the subarachnoid space, the middle one shows that of the z-Evans Index, and the right one shows that of the upper-to-lower subarachnoid space ratio. The black circle points indicate the optimal cutoff points of the maximum area under the ROC curve.
Tables
No. (%) (72 Total) No. (%) (24 Tap-Positive) P Value1a P Value2b No. (%) (25 Tap-Negative) P Value3c No. (%) (23 Controls) Male sex 46 (64%) 15 (62%) 1.000 .917 17 (68%) .831 14 (61%) Evans Index >0.3 42 (58%) 19 (79%) .005d .252 15 (60%) .145 8 (35%) Callosal angle <90° 46 (64%) 21 (88%) <.001d .962 23 (92%) <.001d 2 (9%) High-convexity tightness 35 (49%) 19 (79%) <.001d .154 14 (56%) .002d 2 (9%) Enlarged Sylvian fissure 33 (46%) 19 (79%) <.001d .154 14 (56%) <.001d 0 Alzheimer disease 22 (31%) 4 (17%) .666 .042d 12 (48%) .205 6 (26%) Parkinsonism 5 (7%) 2 (8%) .489 1.000 3 (12%) .263 0 Cerebrovascular diseases 15 (21%) 1 (4%) .022d .115 6 (24%) .615 8 (35%) Narrow spinal canal 7 (10%) 3 (13%) .248 1.000 4 (16%) .139 0 Muscle atrophy of lower leg 12 (17%) 5 (21%) .065 .802 7 (28%) .019d 0 Mean age (± SD) (yr) 76.8 (±6.8) 76.2 (±7.5) .741 .541 77.3 (±5.4) .893 76.8 (±7.8) Mean (72 Total) Mean (24 Tap-Positive) P Value1a P Value2b Mean (25 Tap-Negative) P Value3c Mean (23 Controls) Total ventricle volume (mL) 123 157 <.001d .060 135 <.001d 75.4 Total ventricle volume ratio (%) 8.1 10.5 <.001d .017d 8.8 <.001d 5 Bilateral ventricle volume (mL) 114 147 <.001d .060 125 <.001d 66.9 Bilateral ventricle volume ratio (%) 7.5 9.8 <.001d .013d 8.2 <.001d 4.4 Total subarachnoid space volume (mL) 270 251 .678 .010d 294 .127 264 Total subarachnoid space volume ratio (%) 18.0 16.8 .587 .030d 19.3 .223 17.8 CSF volume of frontal convexity (mL) 51.1 41.0 .003d .016d 52.1 .218 51.1 CSF volume of parietal convexity (mL) 42.4 28.4 <.001d <.001d 44.1 .162 42.4 CSF volume of Sylvian fissure and basal cistern (mL) 119 121 .030d .077 136 <.001d 119 CSF volume of posterior fossa (mL) 57.1 59.1 .232 1.000 58.9 .226 57.1 Upper-to-lower subarachnoid space ratio 0.55 0.37 <.001d .008d 0.49 <.001d 0.79 Evans Index 0.31 0.34 <.001d .028d 0.32 .004d 0.29 Y-Evans Index 0.22 0.23 .001d .053 0.22 .095 0.21 Z-Evans Index 0.38 0.43 <.001d .002d 0.39 <.001d 0.31 X-Max Index 0.63 0.65 <.001d .073 0.64 <.001d 0.60 Y-Max Index 0.63 0.67 <.001d .048d 0.64 .001d 0.58 Z-Max Index 0.70 0.74 <.001d .613 0.74 <.001d 0.62 Callosal angle (degree) 80.9 66.2 <.001d .117 73.4 <.001d 104 Note:—Max indicates maximum.
↵a P value1: probability value for the Mann-Whitney-Wilcoxon test between the tap-positive group and controls.
↵b P value2: probability value for the Mann-Whitney-Wilcoxon test between the tap-positive and tap-negative groups.
↵c P value3: probability value for the Mann-Whitney-Wilcoxon test between the tap-negative group and controls.
↵d Significant.
- Table 3:
At the maximum AUC, thresholds, sensitivity, specificity, and age-adjusted OR (95% CI) for the tap-positive compared with the tap-negative group
Tap-Positive vs Tap-Negative AUC Optimal Threshold Sensitivity Specificity OR (95% CI) P Valuea Total ventricle volume (mL) 0.657 155 58.3 80.0 1.49 (1.13–1.96) .004 Total ventricle volume ratio (%) 0.700 8.6 79.2 56.0 1.49 (1.13–1.96) .005 Bilateral ventricle volume (mL) 0.657 147 54.2 84 1.52 (1.15–2.07) .003 Bilateral ventricle volume ratio (%)b 0.705 7.9 83.3 56.0 1.40 (1.06–1.84) .019 Total subarachnoid space volume (mL)b 0.714 312 87.5 48.0 0.66 (0.49–0.89) .006 Total subarachnoid space volume ratio (%) 0.682 17.9 75.0 72.0 0.63 (0.48–0.82) <.001 CSF volume of frontal convexity (mL)b 0.701 45.9 79.2 76.0 0.58 (0.45–0.74) <.001 CSF volume of parietal convexity (mL)b 0.768 37.9 91.7 52.0 0.60 (0.45–0.79) <.001 CSF volume of Sylvian fissure and basal cistern (mL) 0.648 137 75.0 56.0 0.71 (0.52–0.96) .028 CSF volume of posterior fossa (mL) 0.500 78.7 16.7 96.0 1.41 (0.89–2.25) .146 Upper-to-lower subarachnoid space ratiob 0.723 0.33 45.8 92.0 0.62 (0.46–0.83) .002 Evans Index 0.683 0.33 58.3 80.0 1.37 (1.04–1.80) .026 Y-Evans Index 0.662 0.24 41.7 92.0 1.42 (0.99–2.03) .057 Z-Evans Indexb 0.758 0.42 66.7 88.0 1.74 (1.35–2.24) <.001 X-Max Index 0.640 0.62 70.8 60.0 1.55 (1.12–2.16) .009 Y-Max Index 0.665 0.69 45.8 92.0 1.69 (1.26–2.27) <.001 Z-Max Index 0.543 0.74 58.3 60.0 1.13 (0.85–1.52) .401 Callosal angle (degree) 0.632 77.3 87.5 48.0 1.51 (1.13–2.02) .005
Jump to section
Related Articles
Cited By...
- Assessing CT-based Volumetric Analysis via Transfer Learning with MRI and Manual Labels for Idiopathic Normal Pressure Hydrocephalus
- Prognostic Utility of Disproportionately Enlarged Subarachnoid Space Hydrocephalus in Idiopathic Normal Pressure Hydrocephalus Treated with Ventriculoperitoneal Shunt Surgery: A Systematic Review and Meta-analysis
- Variability of Normal Pressure Hydrocephalus Imaging Biomarkers with Respect to Section Plane Angulation: How Wrong a Radiologist Can Be?
- Comparison of CSF Distribution between Idiopathic Normal Pressure Hydrocephalus and Alzheimer Disease