Deformable multisurface segmentation of the spine for orthopedic surgery planning and simulation.

Purpose: We describe a shape-aware multisurface simplex deformable model for the segmentation of healthy as well as pathological lumbar spine in medical image data. Approach: This model provides an accurate and robust segmentation scheme for the identification of intervertebral disc pathologies to enable the minimally supervised planning and patient-specific simulation of spine surgery, in a manner that combines multisurface and shape statistics-based variants of the deformable simplex model. Statistical shape variation within the dataset has been captured by application of principal component analysis and incorporated during the segmentation process to refine results. In the case where shape statistics hinder detection of the pathological region, user assistance is allowed to disable the prior shape influence during deformation. Results: Results demonstrate validation against user-assisted expert segmentation, showing excellent boundary agreement and prevention of spatial overlap between neighboring surfaces. This section also plots the characteristics of the statistical shape model, such as compactness, generalizability and specificity, as a function of the number of modes used to represent the family of shapes. Final results demonstrate a proof-of-concept deformation application based on the open-source surgery simulation Simulation Open Framework Architecture toolkit. Conclusions: To summarize, we present a deformable multisurface model that embeds a shape statistics force, with applications to surgery planning and simulation.

3D lumbar spine intervertebral disc segmentation and compression simulation from MRI using shape-aware models.

PURPOSE: More accurate and robust image segmentations are needed for identification of spine pathologies and to assist with spine surgery planning and simulation. A framework for 3D segmentation of healthy and herniated intervertebral discs from T2-weighted magnetic resonance imaging was developed that exploits weak shape priors encoded in simplex mesh active surface models. METHODS: Weak shape priors inherent in simplex mesh deformable models have been exploited to automatically segment intervertebral discs. An ellipsoidal simplex template mesh was initialized within the disc image boundary through affine landmark-based registration and was allowed to deform according to image gradient forces. Coarse-to-fine multi-resolution approach was adopted in conjunction with decreasing shape memory forces to accurately capture the disc boundary. User intervention is allowed to turn off the shape feature and guide model deformation when the internal simplex shape memory influence hinders detection of pathology. A resulting surface mesh was utilized for disc compression simulation under gravitational and weight loads using Simulation Open Framework Architecture. For testing, 16 healthy discs were automatically segmented, and five pathological discs were segmented with minimal supervision. RESULTS: Segmentation results were validated against expert guided segmentation and demonstrate mean absolute shape distance error of <1 mm. Healthy intervertebral disc compression simulation resulted in a bulging disc under vertical pressure of 100 N/cm(2). CONCLUSION: This study presents the application of a simplex active surface model featuring weak shape priors for 3D segmentation of healthy as well as herniated discs. A framework was developed that enables the application of shape priors in the healthy part of disc anatomy, with user intervention when the priors were inapplicable. The surface-mesh-based segmentation method is part of a processing pipeline for anatomical modelling to support interactive surgery simulation.

Admission insular infarction >25% is the strongest predictor of large mismatch loss in proximal middle cerebral artery stroke.

BACKGROUND AND PURPOSE: Previous univariate analyses have suggested that proximal middle cerebral artery infarcts with insular involvement have greater severity and are more likely to progress into surrounding penumbral tissue at risk. We hypothesized that a practical, simple scoring method to assess percent insular ribbon infarction (PIRI score) would improve prediction of penumbral loss over other common imaging biomarkers. METHODS: Of consecutive acute stroke patients from 2003 to 2008, 45 with proximal middle cerebral artery-only occlusion met inclusion criteria, including available penumbral imaging. Infarct (diffusion-weighted imaging), tissue at risk (magnetic resonance mean transit time), and final infarct volume (magnetic resonance/computed tomography) were manually segmented. Diffusion-weighted imaging images were rated according to the 5-point PIRI score (0, normal; 1, <25%; 2, 25%-49%; 3, 50%-74%; 4, ≥75% insula involvement). Percent mismatch loss was calculated as an outcome measure of infarct progression. Receiver operating characteristic curve and multivariate analyses were performed. RESULTS: Mean admission diffusion-weighted imaging infarct volume was 30.9 (±38.8) mL and median (interquartile range) PIRI score was 3 (0.75-4). PIRI score was significantly correlated with percent mismatch loss (P<0.0001). When percent mismatch loss was dichotomized based on its median value (30.0%), receiver operating characteristic curve area under curve was 0.89 (P=0.0001) with a 25% insula infarction optimal threshold. After adjusting for time to imaging and treatment, binary logistic regression, including dichotomized PIRI (25% threshold), age, National Institutes of Health Stroke Scale score, diffusion-weighted imaging infarct volume, and computed tomography angiography collateral score as covariates, revealed that only dichotomized insula score (P=0.03) and age (P=0.02) were independent predictors of large (68.2%) versus small (8.1%) mismatch loss. There was excellent interobserver agreement for dichotomized PIRI scoring (κ=0.91). CONCLUSIONS: Admission insular infarction >25% is the strongest predictor of large mismatch loss in this cohort of proximal middle cerebral artery occlusive stroke. This outcome marker may help to identify treatment-eligible patients who are in greatest need of rapid reperfusion therapy.

Diagnostic characteristics of sinonasal organizing hematomas: avoiding misdiagnosis.

BACKGROUND: Organizing hematomas of the paranasal sinuses are diagnostic dilemmas clinically and radiographically, mimicking benign or malignant neoplastic processes and causing patients and clinicians undue worry regarding these diagnoses. Diagnostic criteria for correctly identifying these lesions are not well known. METHODS: A retrospective case series of 7 patients with sinonasal organizing hematoma was studied. Radiographic imaging, clinical characteristics, and pathology were reviewed for new insights. RESULTS: Three patients presented with a primary complaint of epistaxis, 4 had masses visible on nasal endoscopy, and 2 had vascular malformations or small hemangiomas adjacent to the mass found on final pathology. Biopsy of these masses were consistently nondiagnostic prior to complete resection. The most diagnostic findings were "shells" of T2 hypointensity on magnetic resonance imaging (MRI) surrounding the lobules of each of the masses. These correspond to rims of fibrosis at the periphery of the lobules on pathology. Areas of fresh hemorrhage are located at the center of these lobules. CONCLUSION: Sinonasal organizing hematomas are rare lesions of the paranasal sinuses whose clinical characteristics lead to misdiagnoses of benign or malignant neoplasms. Endoscopy, preoperative biopsy, and computed tomography (CT) imaging do not lend helpful information in differentiating these lesions from more worrisome neoplastic processes. However, MRI can lead to positive diagnosis by recognizing the distinct outer rims of T2 hypointensity typically seen in these lesions.