| Purpose: 1/2ABC is the most widely used formula to calculate the volume of intracerebral hemorrhage hematoma,but it often produces a large deviation in the calculation of irregular hematoma and large volume hematoma.In this study,the 3D visual medical image processing platform(3D Slicer)was used as the gold standard to measure hematoma volume,and the existing clinical measurement methods were improved.With the assistance of medical image segmentation algorithm and imaging omics technology,the accuracy of the existing and improved hematoma volume measurement methods was analyzed,so as to find a clinical application.An efficient and accurate method for measuring hematoma volume.Methods: CT image data of 391 patients with intracerebral hemorrhage were collected from the Second Affiliated Hospital of Dalian Medical University from January 2017 to December 2022.The volume of hematoma was measured by 3D Slicer、 s CAVA、 1/2ABC、 1/2ABC1、 1/2ABC2、 2/3SH,respectively(see Appendix 1 for the description of the 6 methods).With the hematoma volume measured by 3D Slicer as the standard,the overall accuracy of the other 5 measurement methods was comprehensively analyzed from the three perspectives of hematoma volume,error value and error rate.At the same time,hematoma volume,hematoma shape and hematoma location were grouped to analyze the accuracy of various measurement methods in different subgroups and the influence of hematoma volume,shape and location on the accuracy of various measurement methods.Results: 1.s CAVA was in good agreement with 3D Slicer in both overall accuracy and subgroup accuracy analysis.2.The overall accuracy of the 5 hematoma volume measurement methods was ranked as follows:s CAVA > 1/2ABC1 > 2/3SH > 1/2ABC > 1/2ABC2.3.Accuracy analysis of 5 methods of hematoma volume measurement in each hematoma volume subgroup.Accuracy ranking for groups less than 10 ml is s CAVA > 2/3SH > 1/2ABC > 1/2ABC1 > 1/2ABC2;The accuracy of 10-30 ml group was as follows: s CAVA > 2/3SH > 1/2ABC1 > 1/2ABC > 1/2ABC2;For groups larger than 30 ml,the accuracy order was s CAVA > 1/2ABC1 > 2/3SH > 1/2ABC > 1/2ABC2.4.There was statistical difference in the 1/2ABC1 error rate between the groups less than 10 ml and more than 30ml(P<0.05),but no statistical difference in the error rate in the other groups(P>0.05).5.Accuracy analysis results of 5 hematoma volume measurement methods in hematoma morphology subgroup: Accuracy ranking of Barras Grade Ⅰ group is s CAVA > 1/2ABC > 2/3SH > 1/2ABC1 > 1/2ABC2;The accuracy ranking of BarrasⅡ and Ⅲ groups is s CAVA>1/2ABC1>2/3SH>1/2ABC>1/2ABC2;The accuracy ranking of BarrasⅣ and Ⅴ groups is s CAVA > 1/2ABC1 > 2/3SH > 1/2ABC > 1/2ABC2。6.A comparative analysis of the differences between BarrasⅣ and Ⅴ patients found that: There was no statistical difference in hematoma volume between 3D Slicer and s CAVA(P=1),no statistical difference in hematoma volume between 3D Slicer and 1/2ABC1(P=1),no statistical difference in hematoma volume,error value and error rate between s CAVA and 1/2ABC1(P=1).The difference test between groups showed that the error rate of 1/2ABC1 had no statistical difference between BarrasⅡ and Ⅲ groups and BarrasⅣ and Ⅴ groups(P=1),while the pairwise comparison of the error rate of other groups had statistical difference(P<0.05).7.Accuracy analysis of 5 hematoma volume measurement methods in hematoma location subgroup.The accuracy of basal ganglia was ranked as s CAVA > 1/2ABC1 > 2/3SH > 1/2ABC > 1/2 ABC2.The accuracy order of non-basal ganglia groups was s CAVA > 1/2ABC1 > 2/3SH > 1/2ABC > 1/2ABC2.The error value and error rate calculated by 1/2ABC1 in the non-basal ganglia area group were lower thanthose in the basal ganglia area.The difference test between groups showed that there were significant differences in both the error value and error rate of 1/2ABC1(P<0.05).Conclusions: 1.s CAVA is an efficient and accurate method for the measurement of hematoma volume,and its accuracy is basically not affected by the size,shape and location of hematoma.2.In the formula method,the irregular hematoma and larger hematoma are more suitable for 1/2ABC1 calculation of hematoma volume,and the accuracy is less affected by the size,shape and location of the hematoma than other formulas.3.In the formula method,hematoma with regular morphology and small volume is more suitable for 1/2ABC calculation of hematoma volume.4.The accuracy of 1/2ABC1 in the calculation of hematoma volume in Barras Ⅳ and Ⅴ patients was comparable to that of 3D Slicer and s CAVA. |
PDF Full Text Request