Differences Between MR Brain Region Segmentation Methods: Impact on Single-Subject Analysis.

For the segmentation of magnetic resonance brain images into anatomical regions, numerous fully automated methods have been proposed and compared to reference segmentations obtained manually. However, systematic differences might exist between the resulting segmentations, depending on the segmentation method and underlying brain atlas. This potentially results in sensitivity differences to disease and can further complicate the comparison of individual patients to normative data. In this study, we aim to answer two research questions: 1) to what extent are methods interchangeable, as long as the same method is being used for computing normative volume distributions and patient-specific volumes? and 2) can different methods be used for computing normative volume distributions and assessing patient-specific volumes? To answer these questions, we compared volumes of six brain regions calculated by five state-of-the-art segmentation methods: Erasmus MC (EMC), FreeSurfer (FS), geodesic information flows (GIF), multi-atlas label propagation with expectation-maximization (MALP-EM), and model-based brain segmentation (MBS). We applied the methods on 988 non-demented (ND) subjects and computed the correlation (PCC-v) and absolute agreement (ICC-v) on the volumes. For most regions, the PCC-v was good ( > 0.75 ), indicating that volume differences between methods in ND subjects are mainly due to systematic differences. The ICC-v was generally lower, especially for the smaller regions, indicating that it is essential that the same method is used to generate normative and patient data. To evaluate the impact on single-subject analysis, we also applied the methods to 42 patients with Alzheimer's disease (AD). In the case where the normative distributions and the patient-specific volumes were calculated by the same method, the patient's distance to the normative distribution was assessed with the z-score. We determined the diagnostic value of this z-score, which showed to be consistent across methods. The absolute agreement on the AD patients' z-scores was high for regions of thalamus and putamen. This is encouraging as it indicates that the studied methods are interchangeable for these regions. For regions such as the hippocampus, amygdala, caudate nucleus and accumbens, and globus pallidus, not all method combinations showed a high ICC-z. Whether two methods are indeed interchangeable should be confirmed for the specific application and dataset of interest.

Low spontaneous variability in cerebral blood flow velocity in non-survivors after cardiac arrest.

OBJECTIVE: To investigate spontaneous variability in the time and frequency domain in mean flow velocity (MFV) and mean arterial pressure (MAP) in comatose patients after cardiac arrest, and determine possible differences between survivors and non-survivors. METHODS: A prospective observational study was performed at the ICU of a tertiary care university hospital in the Netherlands. We studied 11 comatose patients and 10 controls. MFV in the middle cerebral artery was measured with simultaneously recording of MAP. Coefficient of variation (CV) was used as a standardized measure of dispersion in the time domain. In the frequency domain, the average spectral power of MAP and MFV were calculated in the very low, low and high frequency bands. RESULTS: In survivors CV of MFV increased from 4.66 [3.92-6.28] to 7.52 [5.52-15.23] % at T=72h. In non-survivors CV of MFV decreased from 9.02 [1.70-9.36] to 1.97 [1.97-1.97] %. CV of MAP was low immediately after admission (1.46 [1.09-2.25] %) and remained low at 72h (3.05 [1.87-3.63] %) (p=0.13). There were no differences in CV of MAP between survivors and non-survivors (p=0.30). We noticed significant differences between survivors and non-survivors in the VLF band for average spectral power of MAP (p=0.03) and MFV (p=0.003), whereby the power of both MAP and MFV increased in survivors during admission, while remaining low in non-survivors. CONCLUSIONS: Cerebral blood flow is altered after cardiac arrest, with decreased spontaneous fluctuations in non-survivors. Most likely, these changes are the consequence of impaired intrinsic myogenic vascular function and autonomic dysregulation.

Computation of blood flow through collateral circulation of the superficial femoral artery.

OBJECTIVE: Obliteration of collaterals during (endo)vascular treatment of peripheral arterial occlusive disease is considered detrimental. We use a model to calculate maximum collateral bed flow of the superficial femoral artery in order to provide insight in their hemodynamic relevance. METHOD: A computational model was developed using digital subtraction angiographies in combination with Poiseuille's equation and Ohm's law. Lesions were divided into short and long (<15 cm and ≥15 cm, respectively) and into stenosis and occlusions. Data are presented in relation to the calculated maximum healthy superficial femoral artery flow. RESULTS: Stenotic lesions are longer than occlusive lesions (P < 0.05) and occlusions had more and larger collaterals (P < 0.05). In all four study groups the collateral flow significantly increased the total flow (P < 0.05). The maximum collateral system flow in the stenosis and occlusion groups was 5.1% and 20.8% of healthy superficial femoral artery flow, respectively (P < 0.05), and there were no significant differences between short and long lesions (11.2% and 6.7% of healthy superficial femoral artery flow, respectively). CONCLUSION: The maximum collateral system flow of the superficial femoral artery is only a fraction, with a maximum of one fifth, of healthy superficial femoral artery flow. Effects of collateral vessel occlusion during (endo)vascular treatment may therefore be without detrimental consequences.