Limits...
Comparison of the association of sac growth and coil compaction with recurrence in coil embolized cerebral aneurysms.

Hoppe AL, Raghavan ML, Hasan DM - PLoS ONE (2015)

Bottom Line: In recurrent cerebral aneurysms treated by coil embolization, coil compaction is regarded as the presumptive mechanism.The translation of the coil mass center at follow-up was computed.Aneurysm sac growth, not coil compaction, was the primary mechanism of recurrence following successful coil embolization.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Engineering, University of Iowa, Iowa City, Iowa, United States of America.

ABSTRACT

Background and purpose: In recurrent cerebral aneurysms treated by coil embolization, coil compaction is regarded as the presumptive mechanism. We test the hypothesis that aneurysm growth is the primary recurrence mechanism. We also test the hypothesis that the coil mass will translate a measurable extent when recurrence occurs.

Methods: An objective, quantitative image analysis protocol was developed to determine the volumes of aneurysms and coil masses during initial and follow-up visits from 3D rotational angiograms. The population consisted of 15 recurrence and 12 non-recurrence control aneurysms initially completely coiled at a single center. An investigator sensitivity study was performed to assess the objectivity of the methods. Paired Wilcoxon tests (p<0.05, one-tailed) were performed to assess for aneurysm and coil growth. The translation of the coil mass center at follow-up was computed. A Mann Whitney U-Test (p<0.05, one-tailed) was used to compare translation of coil mass centers between recurrence and control subjects.

Results: Image analysis protocol was found to be insensitive to the investigator. Aneurysm growth was evident in the recurrence cohort (p=0.003) but not the control (p=0.136). There was no evidence of coil compaction in either the recurrence or control cohorts (recurrence: p=0.339; control: p=0.429). The translation of the coil mass centers was found to be significantly larger in the recurrence cohort than the control cohort (p=0.047).

Conclusion: Aneurysm sac growth, not coil compaction, was the primary mechanism of recurrence following successful coil embolization. The coil mass likely translates to a measurable extent when recurrence occurs and has the potential to serve as a non-angiographic recurrence marker.

Show MeSH

Related in: MedlinePlus

Schematic illustration of the image processing protocol.‘1- ‘ indicates the pre-first coiling treatment time point; ‘1+’, post-first coiling treatment time point; ‘2-’, pre-second coiling treatment time point; ‘2+’, post-second coiling treatment time point; 3DRA, 3D rotational angiogram. The first column depicts the image processing protocol for the 1- time point, beginning with generation of the aneurysm and vessel model from the subtracted 3DRA scan. The aneurysm sac is then automatically isolated from the vasculature. A representative 1- aneurysm sac model is shown at the bottom of column 1. The adjacent column details the workflow for analyzing the data from the 1+ time point. At this time point the coil mass model is generated from the baseline (or bone) 3DRA scan, while the vessel and residual blood model is generated from the subtracted angiographic scan. The coil mass, vessel and residual blood models are then added together by Boolean union. The aneurysm sac is the combination of the coil mass and any outlying residual blood, of which the aneurysm neck surface is automatically determined. A representative 1+ aneurysm sac model is shown at the bottom of column 2. The remaining two columns outline a similar workflow for analyzing data from the 2- and 2+ time points respectively.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4404091&req=5

pone.0123017.g001: Schematic illustration of the image processing protocol.‘1- ‘ indicates the pre-first coiling treatment time point; ‘1+’, post-first coiling treatment time point; ‘2-’, pre-second coiling treatment time point; ‘2+’, post-second coiling treatment time point; 3DRA, 3D rotational angiogram. The first column depicts the image processing protocol for the 1- time point, beginning with generation of the aneurysm and vessel model from the subtracted 3DRA scan. The aneurysm sac is then automatically isolated from the vasculature. A representative 1- aneurysm sac model is shown at the bottom of column 1. The adjacent column details the workflow for analyzing the data from the 1+ time point. At this time point the coil mass model is generated from the baseline (or bone) 3DRA scan, while the vessel and residual blood model is generated from the subtracted angiographic scan. The coil mass, vessel and residual blood models are then added together by Boolean union. The aneurysm sac is the combination of the coil mass and any outlying residual blood, of which the aneurysm neck surface is automatically determined. A representative 1+ aneurysm sac model is shown at the bottom of column 2. The remaining two columns outline a similar workflow for analyzing data from the 2- and 2+ time points respectively.

Mentions: The image processing algorithms implemented to generate the 3D aneurysm and coil mass models differ slightly depending upon whether the 1- or any of the later time point 3DRA scans (1+, 2-, or 2+) are identified as input. This is because the aneurysm sac is directly visible in the 1- scan but not in the 1+, 2-, or 2+ scans. The 1- aneurysm sac models were each generated from their corresponding subtracted 3DRA image volume because this image features only vessel information. To generate the 3D aneurysm and vessel model, an image segmentation containing both the blood flowing through the contiguous vessels and any blood filling the aneurysm sac was first created. To do this a grayscale morphological opening filter was applied to the subtracted image volume in order to rid it of small-scale noise. Next the user manually places a seed point inside the aneurysm sac; this seed point serves only as an initialization point for the segmentation algorithm and its placement does not affect the segmentation boundaries. Otsu’s algorithm[12] was implemented to generate the segmentation. This algorithm works by finding the statistically optimal threshold between foreground and background voxels in an image volume of interest (VOI).[12] The segmentation was then further evolved using the level set algorithm[13] and the associated 3D model was created from it using the marching cubes algorithm.[14] The aneurysm sac was then isolated from the contiguous vasculature automatically, the result of which was a 3D sac model with a non-planar neck surface per the approach proposed by Ford et al.[15] (see Fig 1). This approach is superior to the use of a single cutting plane to isolate the aneurysm because it retains more of the sac surface features. The 3D models of the 1+, 2-, and 2+ coil masses were each created from the baseline (bone scan) 3DRA image volumes. This was done because the baseline image volumes feature only skull and coil information. A combination of erode and dilate filters were first applied to these image volumes in order to rid the geometry of holes internal to the bounding coil wire. If a hole passed through the entire coil, rendering it a non-simply connected geometry, then this hole was retained. This was done because it was unclear how to fill such a hole in the absence of a bounding coil wire. The resulting coil mass region, representing both the coil wires and interstitial thrombi, was collectively referred to as the coil mass. Similar to the aneurysm and vessel segmentations, Otsu’s segmentation algorithm[12] was then implemented to generate the coil mass segmentation. This segmentation was further evolved using the level set algorithm[13] and the associated 3D model of the coil masses at 1+, 2- and 2+ time periods were created using the marching cubes algorithm.[14] Subsequently, the angiographic volumes (contiguous vessels and residual regions, if any) for 1+, 2- and 2+ time periods were segmented as done for the 1- time period. The aneurysm sac models for 1+, 2- and 2+ were then generated by isolating the sac[15] from the Boolean union of the angiographic volume with the coil mass volume (see Fig 1).


Comparison of the association of sac growth and coil compaction with recurrence in coil embolized cerebral aneurysms.

Hoppe AL, Raghavan ML, Hasan DM - PLoS ONE (2015)

Schematic illustration of the image processing protocol.‘1- ‘ indicates the pre-first coiling treatment time point; ‘1+’, post-first coiling treatment time point; ‘2-’, pre-second coiling treatment time point; ‘2+’, post-second coiling treatment time point; 3DRA, 3D rotational angiogram. The first column depicts the image processing protocol for the 1- time point, beginning with generation of the aneurysm and vessel model from the subtracted 3DRA scan. The aneurysm sac is then automatically isolated from the vasculature. A representative 1- aneurysm sac model is shown at the bottom of column 1. The adjacent column details the workflow for analyzing the data from the 1+ time point. At this time point the coil mass model is generated from the baseline (or bone) 3DRA scan, while the vessel and residual blood model is generated from the subtracted angiographic scan. The coil mass, vessel and residual blood models are then added together by Boolean union. The aneurysm sac is the combination of the coil mass and any outlying residual blood, of which the aneurysm neck surface is automatically determined. A representative 1+ aneurysm sac model is shown at the bottom of column 2. The remaining two columns outline a similar workflow for analyzing data from the 2- and 2+ time points respectively.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4404091&req=5

pone.0123017.g001: Schematic illustration of the image processing protocol.‘1- ‘ indicates the pre-first coiling treatment time point; ‘1+’, post-first coiling treatment time point; ‘2-’, pre-second coiling treatment time point; ‘2+’, post-second coiling treatment time point; 3DRA, 3D rotational angiogram. The first column depicts the image processing protocol for the 1- time point, beginning with generation of the aneurysm and vessel model from the subtracted 3DRA scan. The aneurysm sac is then automatically isolated from the vasculature. A representative 1- aneurysm sac model is shown at the bottom of column 1. The adjacent column details the workflow for analyzing the data from the 1+ time point. At this time point the coil mass model is generated from the baseline (or bone) 3DRA scan, while the vessel and residual blood model is generated from the subtracted angiographic scan. The coil mass, vessel and residual blood models are then added together by Boolean union. The aneurysm sac is the combination of the coil mass and any outlying residual blood, of which the aneurysm neck surface is automatically determined. A representative 1+ aneurysm sac model is shown at the bottom of column 2. The remaining two columns outline a similar workflow for analyzing data from the 2- and 2+ time points respectively.
Mentions: The image processing algorithms implemented to generate the 3D aneurysm and coil mass models differ slightly depending upon whether the 1- or any of the later time point 3DRA scans (1+, 2-, or 2+) are identified as input. This is because the aneurysm sac is directly visible in the 1- scan but not in the 1+, 2-, or 2+ scans. The 1- aneurysm sac models were each generated from their corresponding subtracted 3DRA image volume because this image features only vessel information. To generate the 3D aneurysm and vessel model, an image segmentation containing both the blood flowing through the contiguous vessels and any blood filling the aneurysm sac was first created. To do this a grayscale morphological opening filter was applied to the subtracted image volume in order to rid it of small-scale noise. Next the user manually places a seed point inside the aneurysm sac; this seed point serves only as an initialization point for the segmentation algorithm and its placement does not affect the segmentation boundaries. Otsu’s algorithm[12] was implemented to generate the segmentation. This algorithm works by finding the statistically optimal threshold between foreground and background voxels in an image volume of interest (VOI).[12] The segmentation was then further evolved using the level set algorithm[13] and the associated 3D model was created from it using the marching cubes algorithm.[14] The aneurysm sac was then isolated from the contiguous vasculature automatically, the result of which was a 3D sac model with a non-planar neck surface per the approach proposed by Ford et al.[15] (see Fig 1). This approach is superior to the use of a single cutting plane to isolate the aneurysm because it retains more of the sac surface features. The 3D models of the 1+, 2-, and 2+ coil masses were each created from the baseline (bone scan) 3DRA image volumes. This was done because the baseline image volumes feature only skull and coil information. A combination of erode and dilate filters were first applied to these image volumes in order to rid the geometry of holes internal to the bounding coil wire. If a hole passed through the entire coil, rendering it a non-simply connected geometry, then this hole was retained. This was done because it was unclear how to fill such a hole in the absence of a bounding coil wire. The resulting coil mass region, representing both the coil wires and interstitial thrombi, was collectively referred to as the coil mass. Similar to the aneurysm and vessel segmentations, Otsu’s segmentation algorithm[12] was then implemented to generate the coil mass segmentation. This segmentation was further evolved using the level set algorithm[13] and the associated 3D model of the coil masses at 1+, 2- and 2+ time periods were created using the marching cubes algorithm.[14] Subsequently, the angiographic volumes (contiguous vessels and residual regions, if any) for 1+, 2- and 2+ time periods were segmented as done for the 1- time period. The aneurysm sac models for 1+, 2- and 2+ were then generated by isolating the sac[15] from the Boolean union of the angiographic volume with the coil mass volume (see Fig 1).

Bottom Line: In recurrent cerebral aneurysms treated by coil embolization, coil compaction is regarded as the presumptive mechanism.The translation of the coil mass center at follow-up was computed.Aneurysm sac growth, not coil compaction, was the primary mechanism of recurrence following successful coil embolization.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Engineering, University of Iowa, Iowa City, Iowa, United States of America.

ABSTRACT

Background and purpose: In recurrent cerebral aneurysms treated by coil embolization, coil compaction is regarded as the presumptive mechanism. We test the hypothesis that aneurysm growth is the primary recurrence mechanism. We also test the hypothesis that the coil mass will translate a measurable extent when recurrence occurs.

Methods: An objective, quantitative image analysis protocol was developed to determine the volumes of aneurysms and coil masses during initial and follow-up visits from 3D rotational angiograms. The population consisted of 15 recurrence and 12 non-recurrence control aneurysms initially completely coiled at a single center. An investigator sensitivity study was performed to assess the objectivity of the methods. Paired Wilcoxon tests (p<0.05, one-tailed) were performed to assess for aneurysm and coil growth. The translation of the coil mass center at follow-up was computed. A Mann Whitney U-Test (p<0.05, one-tailed) was used to compare translation of coil mass centers between recurrence and control subjects.

Results: Image analysis protocol was found to be insensitive to the investigator. Aneurysm growth was evident in the recurrence cohort (p=0.003) but not the control (p=0.136). There was no evidence of coil compaction in either the recurrence or control cohorts (recurrence: p=0.339; control: p=0.429). The translation of the coil mass centers was found to be significantly larger in the recurrence cohort than the control cohort (p=0.047).

Conclusion: Aneurysm sac growth, not coil compaction, was the primary mechanism of recurrence following successful coil embolization. The coil mass likely translates to a measurable extent when recurrence occurs and has the potential to serve as a non-angiographic recurrence marker.

Show MeSH
Related in: MedlinePlus