| Objective:Digital software was used to measure the morphological and anatomical parameters of adolescent idiopathic scoliosis in Inner Mongolia,and the differences and correlations among different parameters were analyzed and compared,providing anatomical basis for the selection of treatment methods and preoperative evaluation of the disease.Methods:A total of 300 cases of spinal X-ray images were collected from four grade A hospitals in Inner Mongolia,with no gender limitation and age between 10 and 18 years old.After screening,120 cases with complete image data were finally included as the research object.The image data were transferred to Deepwise workstation(China),the scoliosis morphology and the number of scoliosis were observed,and the anatomical indexes of Cobb Angle,CVA,AVT,TS,CA,CPT,CSI,FPT,CCA,TK,LL,SS,PT and PI were measured.T test and X~2were used to compare the differences of each index in different hospitals,grades,types and genders,and Pearson/Spearman correlation were used to analyze the correlation of each index.Results:There were significant differences in AVT between different scoliosis grades and types(F=34.079,P=0.000;X2=23.379,P=0.000),compared with severe scoliosis,mild and moderate scoliosis AVT was a protective factor,and the smaller the AVT,the less severe the scoliosis.Compared with mild and moderate scoliosis,Cobb Angle of major scoliosis in severe scoliosis was negatively correlated with CCA,LL and SS(r=-0.641,p<0.05;r=-0.695,p<0.01;r=-0.814,p<0.01).There were correlations between LL,SS and PI in different scoliosis indexing.In addition,there were also correlations between CPT,CSI,and CVA in different scoliosis degrees.Conclusion:The coronal and sagittal anatomical parameters of adolescent idiopathic scoliosis in Inner Mongolia are significantly different among different scoliosis grades and types.Compared with mild and moderate scoliosis,more anatomical indicators are significantly correlated in severe scoliosis.Objectives: To develop a deep learning algorithm to automatically evaluate and diagnose scoliosis on full spinal X-ray images.Methods: This retrospective study collected full spinal X-ray images(anteroposterior)from four hospital databases from January 1,2018,to June 30,2021.The data were divided into training and validation sets.Full spinal X-ray images for external validation were independently collected at one hospital from April 1,2021,to June 30,2021.Model effectiveness was validated with a public dataset.Statistical software R was used to analyze the accuracy and sensitivity of the model curvature and anatomical balance parameters and assess interrater consistency.Results: This study included 788 and185 training and test datasets,respectively.The accuracy and recall of the algorithm model for the Cobb angle,apical vertebrae(AV),upper vertebrae,and lower vertebrae were 89.36%,85.71%,77.2%,80.24% and 97.35%,93.38%,84.11%,87.42%,respectively.The symmetric mean absolute percentage error at the Cobb angle was 5.99%,and the automatic measurement time was 1.7 s.The mean absolute error values of the Cobb angle and the distances between the center sacral vertical line and AV and C7 plumb line were 1.07° and 1.12 and 1.38 mm,respectively.Statistical analysis confirmed that the Cobb angle results were in good agreement with the gold standard(interclass coefficients of 0.996,0.978,and 0.825;P<0.001).Conclusion: Our deep learning algorithm model had high sensitivity and accuracy for scoliosis,which could help radiologists improve their diagnostic efficiency. |
PDF Full Text Request