本文运用最小二乘曲线拟合提出一种在医学测量中求解单变量或多变量函数数学模型的方法。它可根据已测得或已知的一组对应函数值,用文中所给出的方程求解函数多项式系数,然后直接写出其表达式。对于在其收敛域内定义的函数,用此方法求得的函数表达式简洁、误差分布均匀一致,可直接用于多个变量和高次函数的曲线拟合。 In this paper, we propose a method to solve the mathematical model of univariate or multivariate functions in medical measurement using least-squares curve fitting. It solves the polynomial coefficients of a function based on a set of corresponding function values ​​that have been measured or known, and then writes out the expression directly. For the function defined in its convergence domain, the function expression obtained by this method is concise, the error distribution is uniform and can be directly used for the curve fitting of multiple variables and high-order functions.