ABSTRACT
<p><b>OBJECTIVE</b>To assess the value of contrast enhanced MRI features for predicting epidermal growth factor receptor () gene amplification in glioblastoma multiforme (GBM) with radiomic method.</p><p><b>METHODS</b>Eighty patients withstatus examined GBM were retrospectively reviewed. The data were randomly divided into a training dataset (60%) and test dataset (40%). Texture features of each case were extracted from the enhanced region and the edema region in contrast enhanced MR images. Principal component analysis was used for dimension reduction. Random forest model, support vector machine model and neural network model were built. Area under the curve (AUC) of the receiver operating characteristics curve was used to assess the performance of models with test dataset.</p><p><b>RESULTS</b>A total of 542 features were extracted from the enhanced region and the edema region. Forty-eight principal components were obtained, which accounted for 100% accumulation contribution rate, and the first 31 principal components were selected for models building, which accounted for 98.5% accumulation contribution rate. The values of AUCs were 0.74, 0.69 and 0.63 for random forest model, support vector machine model and neural network model in the test dataset, respectively.</p><p><b>CONCLUSIONS</b>Radiomic method with proper model may have a potential role in predicting thegene status with enhanced MRI features derived from the enhanced region and the edema region in patients with glioblastoma multiforme.</p>
ABSTRACT
Objective To explore the value of high sensitivity total spine digital radiography in radiation protection and clinical application.Methods A total of 90 patients suffering from scoliosis were randomly divided into three groups with different sensitivity such as S200,S400 and S800.All of them were examined with anterior-posterior and lateral full length spine digital radiography.The S400 group was designated as control group.The doses (mGy) and exposures (mAs) resulted from different groups were calculated and compared,including their means and deviations.Image quality was evaluated with blind method.Results In the S800 group,the doses and exposures were about 43％ and 41％ compared to the S400 group.The differences were statistically significant (t =4.573,8.038,P ＜ 0.05).Doses and exposures in the S200 group were about 187％ and 178％ compared to the control group.The differences were also significant(t =-5.624,-4.052,P ＜ 0.05).In the S200 and S400 group,image quality of anterior-posterior and lateral position was 100％ good.When sensitivity was S800 was selected,image quality of the anterior-posterior position was 100％ good,and 97％ for lateral images.One lateral image was considered noisy.Conclusions In full length spine radiography,high sensitivity (S800) could greatly reduce radiation dose to patients.It is a good choice to use the sensitivity to observe wide-range spine.
ABSTRACT
Objective To study MRI volumetric measurement of hippocampal formation using statistic parametric mapping(SPM) software and to discuss the value of the method applied to Alzheimer's disease (AD).Methods The SPM software was used to divide the three-dimensional MRI brain image into gray matter, white matter and CSF separately.The bilateral hippocampal formations in both AD group and normal control group were delineated and the volumes were measured.The SPM method was compared with conventional method based on region of interest (ROI), which was the gold standard of volume measurement The time used in measuring the volume by these two methods were respectively recorded and compared by two independent samples' t test.Moreover, 7 physicians measured the left hippocampal formation of one same control with both of the two methods.The frequency distribution and dispersion of data acquired with the two methods were evaluated using standard deviation coefficient Results (1)The volume of the bilateral hippocampal formations with SPM method was (1.88±0.07) cm~3 and (1.93±0.08) cm~3 respectively in the AD group, while was (2.99 ±0.07) cm~3 and (3.02 ±0.06) cm~3 in the control group .The volume of bilateral hippocampal formations measured by ROI method was (1.87 ±0.06) cm~3 and (1.91 ±0.09) cm~3 in the AD group, while was (2.97 ±0.08)cm~3 and (3.00 ±0.05) cm~3 in the control group.There was no significant difference between SPM method and conventional ROI method in the AD group and the control group(t=1.500, 1.617, 1.095, 1.889,P> 0.05) .However, the time used for delineation and volume measurement was significantly different.The time used in SPM measurement was (38.1 ±2.0) min, while that in ROI measurement was (55.4 ±2.4) min (t=-25.918, P<0.01).(2)The average volumes of the left hippocampal formation with the two methods measured by the 7 physicians were (2.86±0.20) and (2.76±0.52) cm~3 respectively.The frequency distribution of hippocampal formation volume measured by SPM method and ROI method was different.The CV_(SPM) was 7% and the CV_(ROI) was 19%.Conclusions The borders of hippocampus formation in brain gray matter could be conveniently delineated by SPM software.The accuracy and repetition of measurement was improved by SPM method.The shorter time used in measurement made it possible for MRI volumetric measurement of hippocampal formation to be applied in assessment of relevant neuropsychiatric diseases.
ABSTRACT
Objective To investigate acute CT radiation injury on rabbit liver after the whole body CT scans with high-dose abdominal CT scan model.Methods The forty-six New Zealand white rabbits were randomly assigned into 3 groups.High-dose CT scanning groups included one-off CT scanning group(group 1)and repeated CT scanning group (group 2)with high-dose abdominal CT scan modes,and 18 rabbits in each group.The other two groups were routine-dose CT scanning group(n=5)and control group(n=5).Rats in both group 1 and 2 were divided into 6 subgroups(n= 3).Four to twenty-four times of continuous or discontinuous CT scans with 240 mAs and 140 Kv were performed on every rabbit.The liver samples of every animal were eoUected at 48 h or 24 h after CT scans and pathological examination and immunohistochemistry assay of Cpp32 and Bax expression were performed.Results(1)Light microscope showed①in group 1,slight hepatic injury occurred at 6 times of continuous CT scanning;after 6～18 times of continuous CT scanning,hepatocytes were slight swelling at first,followed by hydropic or fatty degeneration,constriction or emphraxis of hepatic sinuses,puncture necrosis of hepatocytes,and local hepatic hemorrhage;②in group 2:slight hepatic injury occurred at 8 times of CT scanning,and after 8～24 times of CT scanning,hepatocytes were slight or mild swelling at first,followed by moderate or diffuse swelling,hydropic degeneration,and dispersed karyopyknosis of hepatocytes near central vein.(2)Immunohistochemistry assay showed:①in group 1,positive expressions of the apoptosis-inducing protein Bax and Cpp32 were observed in part of liver samples,the expressions of Bax and Cpp32 were correlated with CT radiation dose;②in group 2,positive expressions of the apoptosis-inducing protein Bax and Cpp32 were found in every subgroup,but the expression was not correlated with CT scan times.(3)The hepatic tissues of routine-dose CT scanning group were normal under the light microscope and immunohistochemistry assay.Conclusion Routine-dose CT scanning on liver is very safe,but high-dose CT scanning with non-routine scan times may cause acute pathological hepatic injury and hepatocyte apoptosis.
ABSTRACT
Objective To evaluate the application of latitude reduction technique in CR image processing based on different tissuethickness of foot radiography.Methods The images of adult foot radiography in 50 cases were analysed retrospectively.Latitude reduction technique was used in comparison with routine CR image processing,in order to display different tissue position definition such as soft tissue in phalanges of toes,thumb nail groove,the distal of fifth phalanges,talus bone trabecula and calcaneocuboidal joint space.Results CR image processed by latitude reduction technique could showe clearly tissue and structure of foot and was super than that of routine CR image processing.(t=7.732,P
ABSTRACT
Objective To evaluate the equipment scheme and value of computed radiography (CR) in the fever outpatient department.Methods One scanner of two sets of CR system was placed to the fever outpatient department for obtaining digital images. Those images and relative clinical data were transited to department of radiology by radiological information system (RIS). Then radiologists finished the diagnostic reports and ordered the printer in the fever outpatient department to print after a necessary post-process and analysis for obtained images.Results (1) Only one radiological technician was need to work in the fever outpatient department; (2) The images could preview in 55 seconds and received chest X-ray report in 15 ～ 30 minutes after exposure.Conclusion This project adapts to routine chest X-ray of patients in the fever outpatient department because of preventing contagion and high efficiency.