Deep learning-assisted identification and quantification of aneurysmal subarachnoid hemorrhage in non-contrast CT scans: Development and external validation of Hybrid 2D/3D UNet.

Accurate stroke assessment and consequent favorable clinical outcomes rely on the early identification and quantification of aneurysmal subarachnoid hemorrhage (aSAH) in non-contrast computed tomography (NCCT) images. However, hemorrhagic lesions can be complex and difficult to distinguish manually. To solve these problems, here we propose a novel Hybrid 2D/3D UNet deep-learning framework for automatic aSAH identification and quantification in NCCT images. We evaluated 1824 consecutive patients admitted with aSAH to four hospitals in China between June 2018 and May 2022. Accuracy and precision, Dice scores and intersection over union (IoU), and interclass correlation coefficients (ICC) were calculated to assess model performance, segmentation performance, and correlations between automatic and manual segmentation, respectively. A total of 1355 patients with aSAH were enrolled: 931, 101, 179, and 144 in four datasets, of whom 326 were scanned with Siemens, 640 with Philips, and 389 with GE Medical Systems scanners. Our proposed deep-learning method accurately identified (accuracies 0.993-0.999) and segmented (Dice scores 0.550-0.897) hemorrhage in both the internal and external datasets, even combinations of hemorrhage subtypes. We further developed a convenient AI-assisted platform based on our algorithm to assist clinical workflows, whose performance was comparable to manual measurements by experienced neurosurgeons (ICCs 0.815-0.957) but with greater efficiency and reduced cost. While this tool has not yet been prospectively tested in clinical practice, our innovative hybrid network algorithm and platform can accurately identify and quantify aSAH, paving the way for fast and cheap NCCT interpretation and a reliable AI-based approach to expedite clinical decision-making for aSAH patients.

Endovascular embolization for intracranial aneurysms: report of 162 cases.

OBJECTIVE: to summarize the experiences of endovascular embolization for intracranial aneurysms and emphatically discuss techniques, complications and preventions. METHODS: 171 aneurysms in 162 patients were treated by detachable coil embolization. Among them, 38 cases were treated by GDC, 35 by DCS, and 89 by EDC. RESULTS: 137 aneurysms were 100% occluded, 27 were 90-95% occluded, and 7 were 80% occluded. Complications associated with operation occurred in nine patients. The coil escaping from the sac of aneurysm into its parent artery was seen in five patients. Six patients suffered from rupture of aneurysm during the operation. Serious vasospasm was seen in five patients. Two patients died of complications. CONCLUSIONS: endovascular embolization is a safe, effective and minimally invasive method for treating intracranial aneurysms. Choosing exact embolization techniques is very important to improve the therapeutic effect and decrease the complications of embolization of intracranial aneurysms.

Microsurgical treatment of ruptured intracranial aneurysm: a 120-case analysis.

OBJECTIVE: to investigate the influence of pre-operative conditions and microsurgical skill on the post-operative outcomes of intracranial aneurysms by retrospective analysis of 120 cases with microsurgical treatment. METHODS: 120 patients with 134 intracranial aneurysms received microsurgical treatment via pterional approach or improved pterional approach. RESULTS: of 134 aneurysms, 122 were clipped, one was coated, three were isolated and there was parent artery deligation in one case. 111 Patients were cured, seven cases gave up therapy post-operation, and two died. According to GOS standard, the outcome in the discharge stage was good in 94 cases, mild disability in 12 cases, moderate disability in three cases and severe disability in two cases. Long-term follow-up was performed in all patients, of whom 95 recovered well, mild disability in 12 cases, moderate disability in two cases and severe disability in one case. CONCLUSION: surgical clipping was the most effective method to treat intracranial aneurysm. Optimal chance and microsurgical technique, as well as microanatomical knowledge, are keys for successful treatment.