Objective To evaluate diagnostic accuracy of image interpretation using viewboxes at different luminance with ALVIM statistical phantom (TRG).Methods A digital film of TRG image was obtained by exposing the phantom of TRG with Kodak DR 3000 using proper parameter.Viewboxes with three different luminance were selected from those used in daily work.The film of TRG was evaluated with these viewboxes of three different luminance by three different radiologists with experience of five,eight and sixteen years respectively.A grade scale of 5 was used to evaluate each row of signal and noise.According to the results and formula,the probabilities of true-positive [P (S/s)],false-positive [P (S/n)] and true answer(Pdet) were obtained. Results The three kinds of luminance used in viewboxes were: (3489 ± 256),(2231 ± 220),(816 ± 168) cd/m2.When the average luminance of viewbox was 3489 cd/m2,Pdet of 0.6,0.7,0.8 mm in bone substitute was 0.558±0.009,0.788±0.008,0.813±0.006,Pdet of0.9,1.0,1.2 mm in muscle substitute was 0.663 ± 0.010,0.750 ± 0.008,0.933 ± 0.005 ; When the average luminance of viewbox was 2231 cd/m2 ,Pdet of 0.6,0.7,0.8 mm in bone substitute was 0.525 ± 0.013,0.713 ± 0.013,0.775 ± 0.016,Pdet of 0.9,1.0,1.2 mm in muscle substitute was 0.613 ± 0.019,0.650 ± 0.013,0.850±0.019 ;When the average luminance of viewbox was 816 cd/m2 ,Pdet of 0.6,0.7,0.8 mm in bone substitute was 0.475 ± 0.022,0.550 ± 0.018,0.688±0.020,Pdet of 0.9,1.0,1.2 mm in muscle substitute was 0.550 ± 0.025,0.575 ± 0.021,0.725 ± 0.016. Viewboxes of different luminance were compared to one another by paired t-test.The results of comparison all had significant statistical differences (t = 5.057 and 4.681 ,P <0.05).Conclusions The image displayed by viewboxes at different luminance had significant influence on image interpretation. The higher the luminace of viewbox,the richer the information displayed,and the higher the accuracy of diagnosis.

Objective To compare the differences of metabolic syndrome (MS) prevalence by using four working definitions and their relationship with obesity-related indicators in first-degree relatives of type 2 diabetes mellitus pedigrees. Methods Totally, 2 372 first-degree relatives from 715 type 2 diabetic pedigrees were selected in this study. Complete laboratory data, including blood pressure, lipid profile and plasma glucose, were collected. The prevalence rates of MS and obesity of four definitions, as defined by National Cholesterol Education Program Adult Treatment Panel Ⅲ (ATPⅢ) in 2005, International Diabetes Federation (IDF) in 2005,Chinese Diabetes Society (CDS) in 2004 aml Joint Committee for Developing Chinese Guidelines on Prevention and Treatment of Dyslipidemia in Adults (JCDCG) in 2007,were analyzed. Results (1)The prevalence rates of MS were 45.40% ,38.74% ,25.08% and 39.29% aecording to four definitions respectively. The prevalence rates of MS were higher in females than in males by using ATPⅢ and IDF definitions (both P<0. 01). (2)The common comhinations of metabolic abnormality was dyslipidemia, hypertension, obesity and hyperglycemia by using four definitions,except in females by using CDS definition. (3)The prevalence rates of obesity were 58.18% ,58.18% ,33.90% and 42.96% acconling to the four definitions respectively. The prevalence rates of MS in obese subjects were 66.59% ,66.59% ,54.85% and 68.99% according to four definitions respectively. (4) Applying the cutoff point for abdominal obesity according to ATPⅢ, IDF and JCDCG definitions, the prevalence rates of abdominal obesity in subjects with body mass index (BMI) <25 kg/m2 were respectively 28.58% and 16.78%, being higher in females than in males(38.90% vs 15.02% ,21.01% vs 11.22% ,both P<0. 01). Conclusion (1)There is significant familial aggregation of MS and obesity,and the first-degree relatives of type 2 diabetic patients are high risk populations. (2) Waist circumference rather than BMI taken as a discriminating component of obesity in MS seems to be clinically more helpful to the early identification and prevention of MS.