The Research Of The Application Of Neo Phantom-less QCT In Diagnosis And Prediction Of Osteoporotic Spinal Fractures

Background: With the development of the aging process,the number of osteoporosis patients has ushered in an explosive growth.At present,there are more than 70 million patients in China.Osteoporosis is usually accompanied by a series of complications such as fractures,pain,and reduction of height.Among them,osteoporotic fractures have a high disability rate and fatality rate,which seriously threatens the life and health of the elderly.Osteoporotic vertebral fractures are the most common type of osteoporotic fractures.Because of the lack of obvious incentives and typical symptoms,it is often difficult for patients to self-examine and pay enough attention in the early stage,and most patients have missed the best time for treatment when they seek for medical help.Early screening and intervention for the high-risk groups is an effective method to prevent osteoporotic vertebral fractures.The current clinical risk prediction tools mainly include the area bone mineral density(a BMD)measured by dual energy X-ray absorptiometry(DEXA)and FRAX,but their predictive performance is not so ideal.How to establish an efficient and accurate prediction model for osteoporotic vertebral fractures is a major problem at present.With the continuous maturity of QCT(quantitative computed tomography)technology,many studies have demonstrated its value in diagnosing osteoporosis and predicting osteoporotic fractures.However,there are still many limitations in clinical practice.For this reason,our group and the Hong Kong University team jointly developed an automatic segmentation algorithm for CT imaging and a phantom-less QCT software system,which can be based on different bone structures.The target vertebral body can be automatically divided into cortical bone area and trabecular bone area,and the volumetric bone mineral density parameters of each area are obtained.However,this software system has just come out,and its effectiveness is still in doubt.Therefore,this study used the a BMD measured by DEXA as a reference to verify the accuracy and validity of the software.In addition,this study will establish a phantom-less QCT volume bone mineral density(v BMD)parameter reference centile curve of healthy people in Hunan,and explore the characteristics of each parameter with aging.Futhermore,the prediction and identification efficacy of phantom-less QCT,DEXA and FRAX scores for osteoporotic vertebral fractures were compared,the individual and comprehensive application value of each risk assessment method was explored,in order to find accurate and effective predicting methods for osteoporotic vertebral fractures.Validation of the neo CT imaging automatic segmentation algorithm and phantom-less QCT software systemObjective: 1.Using the DEXA measurement results as a reference,verify whether the measurement results of the CT imaging automatic segmentation algorithm and the phantom-less QCT software system are true and effective.2.Taking vertebral body degeneration as an example,to clarify the limitations of DEXA in the diagnosis of osteoporosis.And to clarify the advantages of phantom-less QCT as a diagnostic tool for osteoporosis.Method: A total of 367 lumbar vertebral segments from 96 subjects were included in this study.First,the DEXA T value of each vertebral body and the v BMD parameter of the phantom-less QCT software system were counted,and all vertebral bodies were divided into osteophyte-positive group and osteophyte-negative group according to lumbar CT images.And the differences in DEXA and phantom-less QCT results of vertebral bodies between the two groups were analyzed.Linear regression analysis and correlation analysis were performed on the a BMD value measured by DEXA of the vertebral body in the osteophyte-negative group and the v BMD parameter measured by the new phantom-less QCT software system to determine the effectiveness of the system.In addition,the distribution pattern of v BMD parameters of vertebral body in osteophyte-positive groups with differences in the diagnosis of DEXA and QCT without phantom was analyzed,and the ROC curve method was used to compare the diagnostic efficacy of DEXA in the diagnosis of osteoporosis.Result:1.Linear regression analysis and correlation analysis were performed on the DEXA a BMD value of the vertebral body in the osteophyte-negative group and the new phantom-less QCT v BMD parameters,and the regression coefficient R2 was 0.73,and the Pearson correlation coefficient was 0.855(P=0.000),showing a significant correlation,confirming that the neo phantom-less QCT is reliable.2.There was a significant difference in the v BMD parameters of the vertebral cortical bone between the osteophyte-positive group and the osteophyte-negative group(263.86mg/cm3 vs 217.63mg/cm3,P=0.000),that is,the presence of vertebral osteophytes greatly strengthened the vertebral body.Cortical bone v BMD parameters of the body,leading to the greater diagnostic deviation of osteoporosis by DEXA.Conclusion: 1.The CT imaging automatic segmentation algorithm and phantom-less QCT software system have reliable measurement results,which can be used as an assistant to the DEXA results to comprehensively evaluate the vertebral bone status.2.Phantom-less QCT is more sensitive than DEXA in evaluating the bone mineral density of patients,especially for patients with osteophytes.3.The presence of spinal osteophytes will increase the volume BMD in the cortical bone area,which will lead to an increase in the area BMD measured by DEXA,thereby increasing the false negative rate of DEXA in diagnosing osteoporosis.Application of lumbar spine volumetric bone mineral density parameter distribution curve in osteoporotic spinal fractures based on the neo phantom-less QCTObjective: 1.Collect spine X-ray images from three medical imaging centers in Changsha,and preliminarily estimate the prevalence of osteoporotic vertebral fractures in people over 50 years old in Hunan.2.Establish the phantom-less QCT volumetric bone mineral density parameter reference centile curve of healthy people in Hunan,and explore the variation rules of each parameter.3.Compare the identification and prediction efficacy of FRAX,DEXA,and phantom-less QCT for osteoporotic vertebral fractures,and explore the value of each risk assessment method alone and in combination,in order to find accurate and effective prediction method for osteoporotic vertebral fractures.Methods: 1.Collect spine X images of a total of 709 subjects over the age of 50 from three imaging centers in Changsha,calculate the prevalence of osteoporotic vertebral fractures in Hunan according to the Genant’s semiquantitative method.2.Adopt a cross-sectional study design,collect CT images of lumbar spine of 442 healthy subjects,and count the v BMD parameters of each vertebral body.Next,established the distribution curve of each v BMD parameter was constructed with age,and found the peak value,trough value,peak age,trough age and variation law of each parameter.In this study,76 patients with confirmed vertebral fractures and 157 healthy control subjects were included,and the differences in v BMD parameters,DEXA T value,FRAX score and height change between the two groups were compared.Logistic regression and ROC curve methods were used to compare the identification and prediction of osteoporotic vertebral fractures when the above indicators were used alone or in combination.Finally,subgroup analysis was performed for the osteoporosis-negative group diagnosed by DEXA and the medium-low fracture risk group by FRAX score.Results: 1.In 2020,the prevalence of osteoporotic vertebral fractures in the middle-aged and elderly population in Hunan was about19.9%,of which the prevalence in males was 12.5% and the prevalence in females was 14.6%.The vertebral body fractures were concentrated in the T11-L1 vertebral bodies.2.All subjects’ trabecular bone v BMD parameters peaked at 20-25 years old,while female subjects’ cortical bone v BMD parameters peaked at around 30 years old,slightly later than that of males at 25 years old.In addition,the rate of change of cortical bone v BMD parameters in all subjects increased significantly after the age of 50,while the loss of trabecular bone structure began in early adulthood,and the rate of metamorphosis was relatively stable.And compared with men,the proportion of bone loss after the age of 50 was significantly higher in female subjects.3.With the basis of traditional osteoporotic vertebral fracture risk assessment tools(such as a BMD parameters of DEXA,FRAX score and height change,etc.),the v BMD parameters of phantom-less QCT(especially the trabecular bone v BMD parameters)can improve the identification and prediction power.In addition,the above analysis was performed on the non-osteoporotic subgroup diagnosed by DEXA and the subgroup judged by the FRAX score as medium and low risk of fracture,and the similar conclusions were obtained.Conclusion: This study is the first to construct a phantom-less QCT volumetric bone mineral density parameter distribution curve of healthy people in Hunan,which suggests the "natural optimal state" of the spine in different age groups.This study found that the vertebral trabecular bone structure in healthy people reached a peak state at the age of 20 to30 years,and then began to decline and the degeneration rate was relatively stable;while the vertebral cortical bone structure peaked relatively late,with a slower decline but a relatively stable rate of degeneration.After 50 years,the degeneration rate increased significantly;and compared with men,the degeneration process of both cortical and trabecular bones in females was significantly increased after 50 years old.This study found that the trabecular bone volume BMD measured by phantom-less QCT can be used as a supplement to traditional osteoporotic fracture risk prediction tools.It has certain significance in identifying and predicting the occurrence of osteoporotic vertebral fractures.However,large sample size clinical study and longitudinal follow-up studies are still needed for further exploration.