Abstract: In the practical application of ultrasonic backscattering measurement,it is essential that how to determine more accurately whether the measured object is osteoporosis.Based on the signal processing results of ultrasonic backscattering experiment,cancellous bone specimens were predicted and classified by supervised support vector machine and adaptive boost algorithms.The results showed that the accuracy of classification was 80.00%-82.86%,and the specificity of osteoporosis was significant(specificity>92.3%).Therefore,the supervised ultrasonic backscattering method is effective and its performance is superior to other existing quantitative ultrasonic methods.It is helpful to the ultrasonic backscattering measurement in vivo.