A Methodology for Systematic Volumetric Analysis of Perioperative Cranial Imaging in Neurosurgical Patients

JHN Journal

Sharan MD, Ashwini²  Wu, MD, MSBmEc, Chengyuan¹  Gorniak, MD, Richard²  Atsina, MD, Kofi-Buaku³ 

[1] Thomas Jefferson University HospitalThomas Jefferson University HospitalThomas Jefferson University Hospital;Thomas Jefferson UniversityThomas Jefferson UniversityThomas Jefferson University;Department of Radiology, Thomas Jefferson University Hospital, Philadelphia PADepartment of Radiology, Thomas Jefferson University Hospital, Philadelphia PADepartment of Radiology, Thomas Jefferson University Hospital, Philadelphia PA

关键词: Image segmentation;    image registration;    image analysis;    workflow;    computer analysis;    JHN Journal;    Thomas Jefferson University Hospital;   

DOI  :  

学科分类：农业科学（综合）

来源: Thomas Jefferson University * Department of Neurological Surgery

【 摘 要 】

Background Although objective assessment of perioperative imaging provides a rigorous evaluation method of neurosurgical techniques in epilepsy, its use remains far from mainstream. Open surgery remains the gold standard for treatment of mesial temporal lobe epilepsy (MTLE); however, stereotactic laser ablation is a promising minimally invasive alternative. Nevertheless, the variables that may affect seizure outcome in stereotactic laser amygdalohippocampectomy (SLAH) remain unclear. While an objective endpoint such as ablated mesial temporal volumes may be significant, a standard methodology for calculating such volumes has yet to be proposed.Objectives To formulate and test a methodology, which can aid in critical evaluation of laser trajectories, and ablation cavities in seizure patients.Methods We performed a retrospective study involving 16 patients undergoing SLAH our institution’s approved IRB protocol. Preoperative MRIs were processed and segmented. Postoperative MRIs were co-registered to preoperative MRIs. Laser trajectories and ablation cavities were segmented from this co-registered image. Segmented trajectories, and cavities were superimposed upon the initial MRI. The percentage of each structure affected was calculated, using a voxel by voxel comparison.Results We were successfully able to determine ablation volumes and critically evaluate laser placement.Conclusion This semi-automated methodology showcases a systematic workflow that objectively evaluates perioperative imaging in neurosurgical patients.Pages: 16-25

【 授权许可】

Unknown