Applicaion Research Of Digital Radiography In Bone Mineral Density Measuring

Objective:With the acceleration of population aging, the incidence rate of osteoporosis has been the seventh kind of common diseasesin the world, osteoporosis （OP） has drawn increasing and a wide range attention of people.Because OP has no obvious clinical symptoms early, but it brings great harm of osteoporotic fracture, so early diagnosis and early treatment is crucial. Bone mineral density （BMD） is an important basis for diagnosis of osteoporosis and evaluating efficacy.Although there are many methods used to determine bone mineral density, in the actual measurement, BMD has certain limitations and many confounding factors （such as the impact of soft tissue, ectopic calcification, bone fracture, radiation is too big, etc.）, It is the urgent need to address important issues how to reduce these influences to improve the precision and accuracy of BMD measurement, in measuring ways of BMD, the dual-energy X ray absorptiometry absorption（DXA） is a recognized "gold standard" internationally to diagnose osteoporosis. But such devices are expensive, so far a total of 350 or so China’s mainland, mostly in large cities, and≥60-year-old group has 130 million elderly people, obviously can not meet the need.Moreover, while DXA measurements in the actual position in the same patients and operating personnel are how to determine the region of interest and other conditions, measurement accuracy are affected; and DXA measurement is not really the size of BMD, bone volume size will be affected; not cancellous bone were measured and the cortical bone, reflecting bone loss and treatment response is not sensitive enough; bone size, shape and other geometric parameters can not be gotten, so DXA is restricted in the diagnosis of osteoporosis and fracture risk prediction.Given the current rapid increase in osteoporosis incidence, DXA was limited in the determination of bone strength and spreading, it is necessary to choose a more practical, simple and accurate technique for measuring BMD and predicting fracture risk. Digital Radiography （DR） technology is a kind of new equipment widely used in clinical imaging diagnosis, not only enriches the form of medical diagnostic information, level, and gradually changed the traditional radiographic techniques and with the rapid development of computer and network technology convergencegreatly improved the quality of medical imaging, while DR has a variety of post-processing system, such as post-processing functions:measurements （size, area, density）, local zoom, contrast, reverse image, image edge enhancement, dual-rate display and subtraction etc, It provide the possibility of BMD measurement, causing the attention of scholars home and abroad.In this study, through the use of DR technology anteroposterior pelvis film, the hip （femoral neck, ward triangle, and greater trochanter） pixel value, corresponding with the DXA BMD measurement were compared, to explore the value of DR technology in BMD measurement.Methods:The Third Hospital of Hebei Medical University, outpatient health check 304 cases, aged 30-70 years old, through interrogation and the necessary laboratory examinations （such as laboratory tests, etc.）, excluding the following:bone tumors and other organic disease,old trauma, secondary osteoporosis, diabetes, a history of endocrine bone diseases, taking drugs affecting bone metabolism recently，liver and kidney dysfunction （alanine aminotransferase> 60U/L, serum creatinine> 133umol/L）and so on. density of the femoral neck, ward triangle and femur greater trochanter three points were determined by developing pelvic plain films according to the standard measurement. At first video technician finished standard DR pelvic plain films, region of interest （ROI） on DR normal pelvic plain film in image post-processing workstation were determined by diagnostic imaging physicians using manual, Choice sites as follows:1. drawing a parallel vertical lines 2cm up the baseline that is connected on the upper edge of both big and small femur trochanter,3 equal parts in the neck area between two lines,3 points is choiced in upper parts; 2. Select ward triangular 1 point; 3. using post-processing workstation mapping tool, select 3 points at about 1 cm2 area in the upper edge of the greater trochanter from the femur 0.5 cm. Femoral neck and greater trochanter were calculated the average of 3 points, and finally come to the pixel density of the femoral neck, ward triangle and greater trochanter, the background value was recorded which there is no organization with the less overlap.BMD of all subjects by dual energy X-ray population absorptiometry were acquired at same period.Stratified by age, database were observed trend, the correlation between two techniques and both side pixel values of DR comparison test were carried out respectively with statistical analysis software.Results:1 normal distribution testmeasured data in various parts are all in line with normal distribution.2 database of DR pixel values and DXA bone mineral density in all ages and all parts of the subjects left femoral neck DR pixels 40ⁱs the highest peak （960.22±281.50） Hu, later declined with age,60ⁱs the lowest （783.19±157.51） Hu; right femoral neck DR pixels 30ⁱs up to the peak （959.97±245.91） Hu, later declined with age,60^to the lowest （806.98±188.65） Hu.left ward triangle DR pixel peak value is around 30-（867.62±213.34） Hu, later declined with age,60^to the lowest （669.52±161.23） Hu; right-ward triangle DR pixel highest peak is 30^{890.46±237.96} Hu, later declined with age,60^to the lowest （700.06±184.34） Hu.on the left greater trochanter, DR pixel is to the highest peak at 40^{696.91±233.42} Hu, later declined with age,60^to the lowest （551.11±154.06） Hu; right side of the greater trochanterDR highest peak pixel is （719.73±236.59） Hu at 40-, later declined with age,60^to the lowest （563.03±156.74） Hu.both femoral neck DXA bone mineral density around the peak were 30-, respectively （0.78±0.12） g/cm2,and （0.79±0.12） g/cm, after declining with age,60^were the lowest （0.68±0.14） g/cm2 and （0.69±0.14） g/cm2.Ward triangle DXA bone mineral density around the peak were 30-, respectively （0.66±0.16） g/cm2,and （0.69±0.17） g/cm2, after declining with age,60～were the lowest （0.50±0.17） g/cm2 and （0.51±0.16） g/cm2.DXA bone mineral density peaks around the greater trochanter were 30～, respectively （0.72±0.11） g/cm2,and （0.69±0.11） g/cm2, after declining with age,60-were the lowest （0.63±0.14）g/cm2 and （0.60±0.13） g/cm2.3 LSD analysis of DR Pixel values and DXA BMD in different parts stratified by ageDR pixel value of bilateral femoral neck:left and right femoral neck 3～had significant increased DR pixel value compared with 50～and 60-respectively （all P<0.01）; 40-had significant increased DR pixel value compared with 50～and 60～respectively （all P<0.01）.DR pixel value of bilateral ward triangle:left and right ward triangle 30-had significant increased DR pixel value compared with 50-and 60 respectively （all P<0.01）; 40-had significant increased DR pixel value compared with 50～and 60～respectively （all P<0.01）.Bilateral greater trochanter DR pixel values:the left greater trochanter 30-had significant increased DR pixel value compared with 50～and 60-respectively （all P<0.01）; 40～had significant increased DR pixel value compared with 50～and 60～respectively （all P<0.01）.Right greater trochanter 30～had significant increased DR pixel value compared with 50～and 60-respectively （P<0.05 and P<0.01）; 40～had significant increased DR pixel value compared with 50～and 60～respectively （all P<0.01）.Bilateral femoral neck DXA bone mineral density:30～DXA bone mineral density in the left femoral neck is significant increased compared with 50～and 60～respectively （all P<0.01）; 40～DXA bone mineral density is significant increased compared with 60-（P<0.01）.Right femoral neck 30-DXA bone mineral density is significant increased compared with 50～and 60～respectively （all P<0.01）; 40～DXA bone mineral density is significant increased compared with 50～and 60～respectively（P<0.05 and P<0.01）.Bilateral ward triangular DXA BMD:the left ward triangle 30～DXA bone mineral density is significant increased compared with 50-and 60-respectively （all P<0.01）;40～had significant increased DXA bone mineral density when compared with 50^and 60-respectively （P<0.05 and P<0.01）; 50^DXA bone mineral density is significant increased compared with the 60-（P<0.05）. right ward triangle 30^DXA bone mineral density is significant increased compared with 50^and 60^respectively （all P<0.01）;40^DXA bone mineral density is significant increased compared with 50^and 60-respectively （all P<0.01）; 50^DXA bone mineral density is significant increased when compared with 60-（P<0.05）.Bilateral greater trochanter DXA bone mineral density:left greater trochanter 30^DXA bone mineral density is significant increased compared with 50^and 60^respectively（P<0.05 and P<0.01）; 40^DXA bone mineral density is significant increased when compared with 60-（P<0.01）.Right greater trochanter 30^DXA bone mineral density is significant increased compared with 50^and 60 respectively （all P<0.01）;40^DXA bone mineral density is significant increased compared with 60-（P<0.01）.4 correlation analysis between DR pixel density and DXA BMD of all parts:left and right femoral neck DXA BMD and DR pixel density showed a significant positive correlation （r= 0.36,0.42） （all P<0.01）; DR pixel density and DXA BMD both ward triangular showed a significant positive correlation （r= 0.515,0.515） （all P<0.01）; DR pixel density and DXA BMD both greater trochanter were positively correlated （r= 0.504,0.464） （all P<0.01）.5 comparison of DR pixels value between left and right sides in all parts: there was no significant difference （t= 1.267, p> 0.05） between the left femoral neck （891.89±94.72） and the right （902.00±92.84） on the DR pixels value; there was no significant difference （t= 1.188, p> 0.05） between the left ward triangle （792.83±87.58） and the right （812.94±76.19） on the DR pixels value; there was no significant difference （t= 1.101, p> 0.05） between the left greater trochanter （635.57±55.99） and the right （655.61±54.81） on pixel value.Conclusion:1 hip peak bone mass in healthy adults is around the ages of 30ⁱn shijiazhung region; 2 hip （double neck, double and two greater trochanter ward triangle） DR pixel value have some differences in each age course, and decreased with age after peak value;3 DR pixel value and DXA BMD between two pairs of femoral neck, ward triangle, and greater tuberosity showed a significant positive correlation, DR pixel values reflect the determination of BMD;4 left and right side in femoral neck, ward triangle, and greater trochanter have no significant difference on DR pixel values, pixel values of the DRvon either side may represent the opposite.