Objective To explore the clinical features of spontaneous spinal epidural hematoma (SSEH) and to find out factors influencing its prognosis. Methods From September 1998 to October 2006, 23 patients with SSEH (10 males and 13 females) were treated. Their ages ranged from 10 to 69 years. The primary neurological status were classified as grade A in 7 patients,B in 2 patients, C in 4 patients, D in 9 patients and E in 1 patients accordingto ASIA grading system. The progressive intervals of their symptoms were divided as four period: less than 12 hours (12 patients), 12 to 24 hours(2 patients), 24 to 48 hours(3 patients) and more than 48 hours(6 patients). SSEH was diagnosedby MRI or by histopathological examination. The cases history, laboratory examination, radiological image, treatment, pathological result and prognosis were recorded and analyzed after 3 month. Results In 23 patients, there were 1 case of deterioration, 8 cases of no change, 9cases of improvement and 5 cases of complete recovery. The gender had no correlationwith prognosis(P＞0.05). In the patients who had shorter progressive interval and more severe edema of spinal cord, the prognosis was worse(P＜0.05). Inthe patients who had mild neurological deficit, the prognosis was good (P＜0.01). In 17 patients undergoing surgery, the scores for prognosis was 1 point in 1 case, 2 points in 5 cases, 3 points in 6 cases and 4 points in 5 cases; the operation time had no correlation with prognosis(r=0.056, P＞0.05). In6patients undergoing conservative treatment, the scores for prognosis were 2 points and 3 points in 3 cases respectively. Conclusion Prognosis of patient with SSEH is influenced by his primary neurological status, progressive interval, spinal edema and size of hematoma. The major treatment is surgical evacuation of hematoma as early as possible to break the aggravation of spinal function. Conservative treatment is not considered unless the neurological defects recovered in the early period.
【Abstract】ObjectiveTo investigate the diagnostic value of the enhancement patterns for characterizing various focal hepatic lesions （FHL）. MethodsForty-seven patients (50 lesions) were included into the study. The morphologic features and the dynamic enhancement patterns of FHL were observed in the early arterial phase, late arterial phase and portal venous phase.The degree of FHL enhancement was analyzed by calculating the contrasttonoise ratio. Results70％ of the HCCs presented “fast-filling and rapid-washout” feature; 67％ of the cholangiocarcinomas showed slight enhancement in arterial phase, and 33.3％ had delayed enhancement on portal venous phase; All hemangiomas presented peripheral nodular enhancement in arterial phase, which then demonstrating centropedal “push-on” enhancement in portal venous phase; Hepatic abscesses mainly presented a slightly enhanced rim around the lesion with fibrous septa inside and an edematous zone outside. ConclusionThe enhancement pattern and the dynamic evolution of FHL enhancement had a great diagnostic value for different FHL by using MRI 3D-VIBE sequence.
One of the main technical challenges when integrating magnetic resonance imaging (MRI) systems with medical linear accelerator is the strong interference of fringe magnetic fields from the MRI system with the electron beams of linear accelerator, making the linear accelerator not to work properly. In order to minimize the interference of magnetic fields, a magnetic shielding cylinder with an open structure made of high permeability materials is designed. ANSYS Maxwell was used to simulate Helmholtz coil which generate uniform magnetic field instead of the fringe magnetic fields which affect accelerator gun. The parameters of shielding tube, such as permeability, radius, length, side thickness, bottom thickness and fringe magnetic fields strength are simulated, and the data is processed by MATLAB to compare the shielding performance. This article gives out a list of magnetic shielding effectiveness with different side thickness and bottom thickness under the optimal radius and length, which showes that this design can meet the shielding requirement for the MRI-linear accelerator system.
ObjectiveTo explore the value of gadobutrol enhanced magnetic resonance angiography (MRA) in abdominal artery angiography.MethodsThe patients were prospectively included for gadobutrol enhanced MRA examination from December 2014 to December 2015. The image quality was assessed by two radiologists. The subjective score and signal intensity were measured for the large and medium arteries, and the subjective score for the small artery was recorded. The Kappa consistency analysis was used to assess the two radiologists’ subjective score.ResultsAll 112 patients were enrolled in this study, 96 of whom were included for the physical examination, 16 of whom were included for the liver tumors. No adverse reactions were found in these patients. The MRA images of 2 patients were affected by the severe respiratory artifact. The MRA images of the other 110 cases were clear and could well show the origins, shapes of large and medium arteries and small arteries. The subjective scores were 21.22±1.93 and 6.24±1.33 of the large and medium arteries and small arteries, respectively. The values of signal noise ratio and contrast signal noise ratio of the large and medium arteries were 1 093.27±331.71 and 897.27±333.29, respectively. The Kappa values of the two radiologists’ subjective score were 0.782 and 0.772 for the large and medium arteries and small arteries, respectively.ConclusionsGadobutrol enhanced MRA can clearly display large and medium arteries, and can also display some small arteries. It has a good application value in abdominal artery angiography.
Focus on the inconsistency of the shape, location and size of brain glioma, a dual-channel 3-dimensional (3D) densely connected network is proposed to automatically segment brain glioma tumor on magnetic resonance images. Our method is based on a 3D convolutional neural network frame, and two convolution kernel sizes are adopted in each channel to extract multi-scale features in different scales of receptive fields. Then we construct two densely connected blocks in each pathway for feature learning and transmission. Finally, the concatenation of two pathway features was sent to classification layer to classify central region voxels to segment brain tumor automatically. We train and test our model on open brain tumor segmentation challenge dataset, and we also compared our results with other models. Experimental results show that our algorithm can segment different tumor lesions more accurately. It has important application value in the clinical diagnosis and treatment of brain tumor diseases.
