放射科网络共享tanner-whitehouse骨龄算法的研究.doc

利用CT图像脂肪自动识别方法进行内脏脂肪检测及肥胖分型 邹诚实作者单位：315300 浙江，慈溪市中医医院放射科 【摘要】 目的 探讨利用CT图像脂肪自动识别方法进行内脏脂肪检测及肥胖分型的可行性。方法 对85例志愿者的肚脐和腰4、5椎体平面进行常规条件CT扫描，所得图像在该方法中按-190Hu～-30Hu的脂肪值范围对腹壁皮下脂肪面积（SA）、腹腔内脏器脂肪面积（VA）、腹围（WC）进行自动识别计算，根据上述计算值及体重指数（BMI）实现肥胖分型，使用SPSS统计学软件分析VA、SA和年龄、WC、BMI的相关性。结果 该方法检测内脏脂肪面积方法简便，准确性及自动化程度高，能够即时得出肥胖类型（正常、皮下脂肪型肥胖、内脏脂肪型蓄积、内脏脂肪型肥胖）。VA和WC、BMI显著性相关，而与年龄弱相关。SA和WC、BMI显著性相关，而与年龄无明显相关性。VA和SA也显著性相关。结论 利用CT图像脂肪自动识别方法进行内脏脂肪检测及肥胖分型切实可行。 【关键词】　腹部脂肪　断层摄影术；X线计算机　测量；计算机辅助 Measure visceral adipose area and classify obesity with automatically identify the abdominal adipose by CT images ZOU Cheng-shi. Department of Radiology,Cixi TCM Hospital,Cixi 315300,Zhejiang Province,China 【Abstract】Objective To evaluate the feasibility of measurement of visceral adipose area and classification of obesity with automation identification technology of the abdominal adipose by CT images. Methods The navel and the 4-5 lumbar vertebrae of 85 volunteers was scanned by CT under general condition, abdominal subcutaneous adipose area (SA) and visceral adipose area (VA) and waist circumference (WC) are calculated and distinguished by-190Hu~-30Hu adipose value range. the classification of obesity in accordance with the calculation and BMI, and analysis the correlation between in the VA, SA and age, WC, BMI using SPSS. Results This method of measurement of visceral adipose area is easy, accurate, and have high degree of automation. The classification of obesity (Normal, Subcutaneous adipose type, Visceral adipose accumulation, Visceral adipose type) can be obtained at once. Significant correlation not only exists between VA and WC or BMI but also exists between SA and WC or BMI, a slight correlation exists between VA and age, not between in SA and age. Significant correlations also exist between VA and SA. Conclusion The method of measurement of visceral adipose area and classification of obesity Automation Identification Technology of the abdominal adipose by CT images is feasible. 【Key words】　Abdominal adip