There were 225 (70 %) benign and 99 (30 %) malignant ovarian masses. Using logistic regression with the area under the curve of the receiver operating curve of 82 %, the cancer probability was determined by the equation.
$$ \frac{{{\text{e}}^{{ - 3. 6 3 7 2 { } + \, 0.0 30 6 { }* \, \left( {\text{A}} \right) \, + \, 0.00 1 { }* \, \left( {\text{B}} \right) \, + \, 0.{ 876 }* \, \left( {\text{C}} \right) + 1. 5 5 1 { }* \, \left( {\text{D}} \right) \, + { 1}. 7 3 7 7 { }* \, \left( {\text{E}} \right) \, + { 2}. 7 6 { }* \, \left( {\text{F}} \right)}} }}{{ 1+ {\text{e}}^{{ - 3. 6 3 7 2 { } + \, 0.0 30 6 { }* \, \left( {\text{A}} \right) \, + \, 0.00 1 { }* \, \left( {\text{B}} \right) \, + \, 0. 8 7 6 { }* \, \left( {\text{C}} \right) \, + { 1}. 5 5 1 { }* \, \left( {\text{D}} \right) \, + { 1}. 7 3 7 7 { }* \, \left( {\text{E}} \right) \, + { 2}. 7 6 { }* \, \left( {\text{F}} \right)}} }} $$
where A = age, B = CA-125, C = solid adnexal mass is 1 and cystic is 0, D = ascites is 1, E = omental caking is 1 and absence is 0, F = node size ≥1 cm is 1 and <1 cm is 0 value. The natural logarithm e is a constant [2.718281828]. For example, for a woman of age 60, CA-125 = 50 U/mL, with solid adnexal mass, ascites, omental caking, and lymphadenopathy, the probability is 0.994. Hence, this woman has a 99.4 % probability of having cancer.