Provided are an apparatus and a computer-readable storage medium having stored therein a program that can determine keratoconus with a simple configuration so as to allow the apparatus and the program to be distributed widely and contribute to early diagnosis of keratoconus.In a keratometer as a keratoconus determination apparatus, a prediction model for keratoconus is stored in a memory. The prediction model is a logistic regression model in which three parameters that are a steep meridian refractive power, a flat meridian refractive power, and a value indicating whether or not a subject eye has a with-the-rule astigmatism, are independent variables, and a probability of keratoconus is a dependent variable. A control unit substitutes the three parameter values into the prediction model, to obtain a probability of keratoconus. When the probability is greater than a cutoff value, the subject eye is determined to be suspected of having keratoconus.
