Comparison of computed tomography angiography with digital subtraction angiography in the assessment of clipped intracranial aneurysms

Automated detection of intracranial aneurysms using skeleton-based 3D patches, semantic segmentation, and auxiliary classification for overcoming data imbalance in brain TOF-MRA

Accurate and reliable detection of intracranial aneurysms is vital for subsequent treatment to prevent bleeding. However, the detection of intracranial aneurysms can be time-consuming and even challenging, and there is great variability among experts, especially in the case of small aneurysms. This study aimed to detect intracranial aneurysms accurately using a convolutional neural network (CNN) with 3D time-of-flight magnetic resonance angiography (TOF-MRA). A total of 154 3D TOF-MRA datasets with intracranial aneurysms were acquired, and the gold standards were manually drawn by neuroradiologists. We also obtained 113 subjects from a public dataset for external validation. These angiograms were pre-processed by using skull-stripping, signal intensity normalization, and N4 bias correction. The 3D patches along the vessel skeleton from MRA were extracted. Values of the ratio between the aneurysmal and the normal patches ranged from 1:1 to 1:5. The semantic segmentation on intracranial aneurysms was trained using a 3D U-Net with an auxiliary classifier to overcome the imbalance in patches. The proposed method achieved an accuracy of 0.910 in internal validation and external validation accuracy of 0.883 with a 2:1 ratio of normal to aneurysmal patches. This multi-task learning method showed that the aneurysm segmentation performance was sufficient to be helpful in an actual clinical setting.

Follow-up imaging of clipped intracranial aneurysms with 3-T MRI: comparison between 3D time-of-flight MR angiography and pointwise encoding time reduction with radial acquisition subtraction-based MR angiography

OBJECTIVE

Metallic susceptibility artifact due to implanted clips is a major limitation of using 3D time-of-flight magnetic resonance angiography (TOF-MRA) for follow-up imaging of clipped aneurysms (CAs). The purpose of this study was to compare pointwise encoding time reduction with radial acquisition (PETRA) subtraction-based MRA with TOF-MRA in terms of imaging quality and visibility of clip-adjacent arteries for use in follow-up imaging of CAs.

METHODS

Sixty-two patients with 73 CAs were included retrospectively in this comparative study. All patients underwent PETRA-MRA after TOF-MRA performed simultaneously with 3-T MRI between September 2019 and March 2020. Two neuroradiologists independently compared images obtained with both MRA modalities to evaluate overall image quality using a 4-point scale and visibility of the parent artery and branching vessels near the clips using a 3-point scale. Subgroup analysis was performed according to the number of clips (less-clipped [1-2 clips] vs more-clipped [≥ 3 clips] aneurysms). The ability to detect aneurysm recurrence was also assessed.

RESULTS

Compared with TOF-MRA, PETRA-MRA showed acceptable image quality (score of 3.97 ± 0.18 for TOF-MRA vs 3.73 ± 0.53 for PETRA-MRA) and had greater visibility of the adjacent vessels near the CAs (score of 1.25 ± 0.59 for TOF-MRA vs 2.27 ± 0.75 for PETRA-MRA, p < 0.0001). PETRA-MRA had greater visibility of vessels adjacent to less-clipped aneurysms (score of 2.39 ± 0.75 for less-clipped aneurysms vs 2.09 ± 0.72 for more-clipped aneurysms, p = 0.014). Of 73 CAs, aneurysm recurrence in 4 cases was detected using PETRA-MRA.

CONCLUSIONS

This study demonstrated that PETRA-MRA is superior to TOF-MRA for visualizing adjacent vessels near clips and can be an advantageous alternative to TOF-MRA for follow-up imaging of CAs.

Intraoperative fluorescein video angiography in intracranial aneurysm surgery: single-center, observational cohort study

The aim of this study was to compare the incidence of postoperative, surgery-related complications in patients where fluorescein video angiography (FL-VA) was performed with those operated without intraoperative verification. This is an observational cohort study including 97 patients who were selected for microsurgical clipping due to intracranial aneurysm. First 52 patients enrolled in the study were operated prior to introduction of fluorescein fluorescence in our surgical workflow. These patients were considered as controls. The study group consisted of 45 consecutive patients operated with the use of fluorescein video angiography and by the same surgical team. Outcomes in both groups were compared using non-parametric test (Mann-Whitney U). Intraoperative fluorescein video angiography revealed aneurysm remnant or inadvertent vessel occlusion in 17.8% of patients. Following clip reposition, a repeated FL-VA was performed to confirm restoration of blood flow and/or complete aneurysm obliteration. Intraoperative findings were later confirmed using computed tomography angiography (CTA). None of the patients in our study group developed surgery-related complications; whereas in the control group, aneurysm remnant was discovered in 7.7%, brain ischemia in 9.6% and both of the latter in 5.8% of patients. Difference in treatment-related outcome was statistically significant (p < 0.05). Intraoperative fluorescein video angiography successfully identified aneurysm residual and adjacent artery occlusion leading to excellent outcome following clip reposition.

Frequency and risk factors for postoperative aneurysm residual after microsurgical clipping

OBJECTIVE

Aneurysm residuals after clipping are a well-known problem, but the course of aneurysm remnants in follow-up is not well studied. No standards or follow-up guidelines exist for treatment of aneurysm remnants. The aim of this study was to evaluate the risk factors for postoperative aneurysm remnants and their changes during follow-up.

METHODS

We performed a retrospective analysis of 666 aneurysms treated via clipping in our hospital from 2006 to 2016. Postoperative and follow-up angiographic data were analyzed for aneurysm remnants and regrowth. Clinical parameters and aneurysm-specific characteristics were correlated with radiological results.

RESULTS

The frequency of aneurysm residuals was 12% (78/666). Aneurysms located in the middle cerebral artery (p = 0.02) showed a significantly lower risk for incomplete aneurysm occlusion. Larger aneurysms with a diameter of 11-25 mm (p = 0.005) showed a significantly higher risk for incomplete aneurysm occlusion. Five patients underwent re-clipping during the same hospital stay. Remnants were stratified based on morphological characteristics into "dog ears" (n = 60) and "broad based" (n = 13). The majority of the "dog ears" stayed stable, decreased in size, or vanished during follow-up. Broad-based remnants showed a higher risk of regrowth.

CONCLUSIONS

A middle cerebral artery location seems to lower the risk for the incomplete clip occlusion of an aneurysm. Greater aneurysm size (11-25 mm) is associated with a postoperative aneurysm remnant. The majority of "dog-ear" remnants appear to remain stable during follow-up. In these cases, unnecessarily frequent angiographic checks could be avoided. By contrast, broad-based residuals show a higher risk of regrowth that requires close imaging controls if retreatment cannot be performed immediately.

256-row multislice CT angiography in the postoperative evaluation of cerebral aneurysms treated with titanium clips: using three-dimensional rotational angiography as the standard of reference

OBJECTIVE

To assess the diagnostic accuracy of 256-row multislice computed tomographic angiography (CTA) compared with three-dimensional rotational angiography (3DRA) in the postoperative evaluation of cerebral aneurysms treated with titanium clips.

METHODS

A total of 128 patients (42 men, 86 women; mean age, 57.6 years) with 143 cerebral aneurysms treated using titanium clips underwent both CTA and 3DRA. Two reviewers retrospectively evaluated the following parameters on CTA and 3DRA: (1) residual/recurrent aneurysm (absent or present), (2) patency of parent artery (patent or occluded/severe stenotic (> 70%)), and (3) patency of adjacent branch (patent or occluded/absent).

RESULTS

A total of 24 residual/recurrent aneurysms were detected by 3DRA. The sensitivity, specificity, and accuracy of CTA for the detection of residual/recurrent aneurysms were 83.3%, 100%, and 97.2% for reviewer 1 and 79.2%, 100%, and 96.5% for reviewer 2, respectively. The sensitivity, specificity, and accuracy of CTA for the evaluation of patency of parent artery were 100%, 100%, and 100%, respectively, for both reviewers. The sensitivity, specificity, and accuracy of CTA for evaluation of the patency of adjacent branch were 85.1%, 100%, and 92.3% for reviewer 1 and 82.4%, 100%, and 90.9% for reviewer 2, respectively.

CONCLUSION

A 256-row multislice CTA is a valuable non-invasive tool for assessment of cerebral aneurysms treated with titanium clips.

KEY POINTS

• A 256-row multislice CTA is an accurate imaging technique for the postoperative assessment of cerebral aneurysms treated with titanium clips. • Sensitivity of CTA for the detection of residual/recurrent aneurysms was 79-83% compared with 3DRA. • CTA is still limited in detecting residual/recurrent aneurysms of < 2 mm and small adjacent branches.

Deep Learning-Assisted Diagnosis of Cerebral Aneurysms Using the HeadXNet Model

IMPORTANCE

Deep learning has the potential to augment clinician performance in medical imaging interpretation and reduce time to diagnosis through automated segmentation. Few studies to date have explored this topic.

OBJECTIVE

To develop and apply a neural network segmentation model (the HeadXNet model) capable of generating precise voxel-by-voxel predictions of intracranial aneurysms on head computed tomographic angiography (CTA) imaging to augment clinicians' intracranial aneurysm diagnostic performance.

DESIGN, SETTING, AND PARTICIPANTS

In this diagnostic study, a 3-dimensional convolutional neural network architecture was developed using a training set of 611 head CTA examinations to generate aneurysm segmentations. Segmentation outputs from this support model on a test set of 115 examinations were provided to clinicians. Between August 13, 2018, and October 4, 2018, 8 clinicians diagnosed the presence of aneurysm on the test set, both with and without model augmentation, in a crossover design using randomized order and a 14-day washout period. Head and neck examinations performed between January 3, 2003, and May 31, 2017, at a single academic medical center were used to train, validate, and test the model. Examinations positive for aneurysm had at least 1 clinically significant, nonruptured intracranial aneurysm. Examinations with hemorrhage, ruptured aneurysm, posttraumatic or infectious pseudoaneurysm, arteriovenous malformation, surgical clips, coils, catheters, or other surgical hardware were excluded. All other CTA examinations were considered controls.

MAIN OUTCOMES AND MEASURES

Sensitivity, specificity, accuracy, time, and interrater agreement were measured. Metrics for clinician performance with and without model augmentation were compared.

RESULTS

The data set contained 818 examinations from 662 unique patients with 328 CTA examinations (40.1%) containing at least 1 intracranial aneurysm and 490 examinations (59.9%) without intracranial aneurysms. The 8 clinicians reading the test set ranged in experience from 2 to 12 years. Augmenting clinicians with artificial intelligence-produced segmentation predictions resulted in clinicians achieving statistically significant improvements in sensitivity, accuracy, and interrater agreement when compared with no augmentation. The clinicians' mean sensitivity increased by 0.059 (95% CI, 0.028-0.091; adjusted P = .01), mean accuracy increased by 0.038 (95% CI, 0.014-0.062; adjusted P = .02), and mean interrater agreement (Fleiss κ) increased by 0.060, from 0.799 to 0.859 (adjusted P = .05). There was no statistically significant change in mean specificity (0.016; 95% CI, -0.010 to 0.041; adjusted P = .16) and time to diagnosis (5.71 seconds; 95% CI, 7.22-18.63 seconds; adjusted P = .19).

CONCLUSIONS AND RELEVANCE

The deep learning model developed successfully detected clinically significant intracranial aneurysms on CTA. This suggests that integration of an artificial intelligence-assisted diagnostic model may augment clinician performance with dependable and accurate predictions and thereby optimize patient care.

Residual Cerebral Aneurysms After Microsurgical Clipping: A New Scale, an Agreement Study, and a Systematic Review of the Literature

OBJECTIVE

The surgical repair of a cerebral aneurysm does not always lead to complete occlusion. A standardized repeatable method of reporting results of surgical clipping is desirable. Our purpose was to systematically review methods of classifying aneurysm remnants, provide a new scale with precise definitions of categories, and perform an agreement study to assess the variability in adjudicating remnants after aneurysm clipping.

METHODS

A systematic review was performed to identify ways to report angiographic results of surgical clipping between 1963 and 2017. Postclipping angiographic results of 43 patients were also independently evaluated by 10 raters of various experience and backgrounds using a new 4-category scale. Agreement between responses were analyzed using κ statistics.

RESULTS

The systematic review yielded 63 articles with 37 different nomenclatures using 2-6 categories. The reliability of judging the presence of an aneurysm remnant on catheter angiography was studied only twice, with only 2 raters each time, with contradictory results. Interobserver agreement using the new 4-category scale was moderate (κ = 0.52; 95% confidence interval, 0.43-0.62) for all observers, but improved to substantial (κ = 0.62; 95% confidence interval, 0.47-0.76) when results were dichotomized (grade 0/1 vs. 2/3).

CONCLUSIONS

Various classification schemes to evaluate angiographic results after surgical clipping exist in the literature, but they lack standardization. Adjudication using fewer, better defined categories may yield more reliable agreement.

The utility of dual-energy CT for metal artifact reduction from intracranial clipping and coiling

OBJECTIVE

To assess the ability of dual-energy CT (DECT) to reduce metal-related artifacts in patients with clips and coils in head CT angiography, and to analyze the differences in this reduction between both type of devices.

MATERIALS AND METHODS

Thirteen patients (6 clips, 7 coils) were selected and retrospectively analized. Virtual monoenergetic images (MEI) with photon energies from 40 to 150 keV were obtained. Noise was measured at the area of maximum artifact. Subjective evaluation of streak artifact was performed by two radiologists independently. Differences between noise values in all groups were tested by using the ANOVA test. Mann-Whitney U test was used to compare the differences between clips and coils. Coheńs κ statistic was used to determine interobserver agreement.

RESULTS

The lowest noise value was observed at high energy levels (p<0,05). Noise was higher in the coil group than in the clip group (p<0.001). Interobserver agreement was good (κ=0.72).

CONCLUSIONS

TCED with MEI helps to minimize the artifact from clips ands coils in patients who undergo head CT angiography. The reduction of the artifact is greater in patients with surgical clipping than in patients with endovascular coiling.

Reduction of Metallic Artifacts of the Post-treatment Intracranial Aneurysms: Effects of Single Energy Metal Artifact Reduction Algorithm

PURPOSE

This study evaluated the quality of computed tomography (CT) and CT angiography images generated using the single-energy metal artifact reduction (SEMAR) algorithm during perfusion examination in patients who had undergone reconstruction with neurosurgical clipping or endovascular coiling for treatment of aneurysms.

METHODS

A total of 55 patients with implanted intracranial clips or coils (24 men and 31 women; mean age 60.15 ± 15.86 years) underwent perfusion studies evaluated by CT and CT angiography with a 320-row CT scanner. Images were reconstructed with either the SEMAR algorithm combined with iterative reconstruction (SEMAR group), or by iterative reconstruction only (non-SEMAR group control). The SEMAR and control images were compared for artifacts (index and maximum diameter), image quality, cerebral perfusion parameters, noise (images with the worst artifacts), and contrast-to-noise ratio. The metallic artifacts were visually evaluated by two radiologists using a four-point scale in a double-blinded manner.

RESULTS

The noise, artifact diameter, and artifact index of the SEMAR images were significantly lower than that of the control images, and the subjective image quality score and contrast-to-noise ratio were significantly higher (P < 0.01, all). The cerebral perfusion parameters of the SEMAR and control images were comparable (i. e. blood flow, blood volume, and mean transit time).

CONCLUSION

For imaging intracranial metallic implants, the SEMAR algorithm produced images with significantly fewer artifacts than the iterative reconstruction alone, with no statistical changes in perfusion parameters. Thus, SEMAR reconstruction can be instrumental in improving CT image quality and may ultimately improve the detection of postoperative complications and patient prognosis.

Current Treatment Strategies for Intracranial Aneurysms: An Overview

Intracranial aneurysm is a leading cause of stroke. Its treatment has evolved over the past 2 decades. This review summarizes the treatment strategies for intracranial aneurysms from 3 different perspectives: open surgery approach, transluminal treatment approach, and new technologies being used or trialed. We introduce most of the available treatment techniques in detail, including contralateral clipping, wrapping and clipping, double catheters assisting coiling and waffle-cone technique, and so on. Data from major trials such as Analysis of Treatment by Endovascular approach of Non-ruptured Aneurysms (ATENA), Internal Subarachnoid Trial (ISAT), Clinical and Anatomical Results in the Treatment of Ruptured Intracranial Aneurysms (CLARITY), and Barrow Ruptured Aneurysm Trial (BRAT) as well as information from other clinical reports and local experience are reviewed to suggest a clinical pathway for treating different types of intracranial aneurysms. It will be a valuable supplement to the current existing guidelines. We hope it could help assisting real-time decision-making in clinical practices and also encourage advancements in managing the disease.

Aneurysmal subarachnoid haemorrhage from a neuroimaging perspective

Neuroimaging is a key element in the management of patients suffering from subarachnoid haemorrhage (SAH). In this article, we review the current literature to provide a summary of the existing neuroimaging methods available in clinical practice. Noncontrast computed tomography is highly sensitive in detecting subarachnoid blood, especially within 6 hours of haemorrhage. However, lumbar puncture should follow a negative noncontrast computed tomography scan in patients with symptoms suspicious of SAH. Computed tomography angiography is slowly replacing digital subtraction angiography as the first-line technique for the diagnosis and treatment planning of cerebral aneurysms, but digital subtraction angiography is still required in patients with diffuse SAH and negative initial computed tomography angiography. Delayed cerebral ischaemia is a common and serious complication after SAH. The modern concept of delayed cerebral ischaemia monitoring is shifting from modalities that measure vessel diameter to techniques focusing on brain perfusion. Lastly, evolving modalities applied to assess cerebral physiological, functional and cognitive sequelae after SAH, such as functional magnetic resonance imaging or positron emission tomography, are discussed. These new techniques may have the advantage over structural modalities due to their ability to assess brain physiology and function in real time. However, their use remains mainly experimental and the literature supporting their practice is still scarce.

Recent advances in diagnostic approaches for sub-arachnoid hemorrhage

Sub-arachnoid hemorrhage (SAH) has been easily one of the most debilitating neurosurgical entities as far as stroke related case mortality and morbidity rates are concerned. To date, it has case fatality rates ranging from 32-67%. Advances in the diagnostic accuracy of the available imaging methods have contributed significantly in reducing morbidity associated with this deadly disease. We currently have computed tomography angiography (CTA), magnetic resonance angiography (MRA) and the digital subtraction angiography (DSA) including three dimensional DSA as the mainstay diagnostic techniques. The non-invasive angiography in the form of CTA and MRA has evolved in the last decade as rapid, easily available, and economical means of diagnosing the cause of SAH. The role of three dimensional computed tomography angiography (3D-CTA) in management of aneurysms has been fairly acknowledged in the past. There have been numerous articles in the literature regarding its potential threat to the conventional “gold standard” DSA. The most recent addition has been the introduction of the fourth dimension to the established 3D-CT angiography (4D-CTA). At many centers, DSA is still treated as the first choice of investigation. Although, CT angiography still has some limitations, it can provide an unmatched multi-directional view of the aneurysmal morphology and its surroundings including relations with the skull base and blood vessels. We study the recent advances in the diagnostic approaches to SAH with special emphasis on 3D-CTA and 4D-CTA as the upcoming technologies.

Diagnostic accuracy of early computed tomographic angiography for visualizing medium sized inferior and posterior projecting carotid system aneurysms

Background

Conventional angiography, generally referred to as intra-arterial digital subtraction angiography, still remains the gold standard reference method for the diagnosis of intracranial aneurysms, helical computed tomography angiography (CTA) is a new non-invasive volumetric imaging method.

Objectives

This study was conducted to screen patients presenting with subarachnoidhemorrhage by CTA before conventional digital subtraction angiography (DSA) and subsequently comparing the results for various aneurysm projections.

Patients and Methods

In a prospective study, 99 consecutive patients with an initial diagnosis of subarachnoid hemorrhage were screened for aneurysms with CTA followed by conventional DSA. There were 17 cases with negative angiograms in whom repeat angiograms, three months later were negative for 15 cases, while two cases were found to bear aneurysm on the repeat examination. Eighty two patients had at least one proven aneurysm on initial DSA and two on the repeat angiogram. Out of 84 patients, five underwent endovascular treatment and 79 patients who underwent surgical clipping were considered for projection evaluation.

Results

Sensitivity of CTA was 98.78% (95% confidence interval [CI], 93.4-99.7%), while the specificity was 100% (95% CI, 81.57-100%) and the kappa coefficient of agreement between CTA and DSA was 96.5%. The most significant discrepancies with DSA findings were for visualizing the projection of inferior and posterior projecting proximal anterior circulation aneurysms.

Conclusions

Helical CTA was in good concordance with DSA for screening of cerebral aneurysms; however, for exact visualization of the aneurysm neck and its projection, especially if it is inferior or posterior, DSA remains the gold standard.

Current imaging assessment and treatment of intracranial aneurysms

OBJECTIVE

This article reviews current neuroimaging techniques used for screening, diagnosis, and follow-up of patients with intracranial aneurysms as well as neuroendovascular therapeutic options available to patients.

CONCLUSION

The diagnosis and management of intracranial aneurysms have evolved dramatically in the past 20 years. MR angiography and CT angiography allow radiologists to reliably and noninvasively diagnose most intracranial aneurysms. Nonoperative endovascular techniques for treating intracranial aneurysms are now making treatment increasingly safer and more effective.