Description

Timmerman et al developed 2 logistic regression models for evaluating a woman with an adnexal mass prior to surgery. The simplified model M2 has half the parameters of the more complex model M1. The authors are from multiple university hospitals in Europe representing the International Ovarian Tumor Analysis Group.


 

Parameters:

(1) age of the patient in years

(2) presence of ascites

(3) presence of blood flow within a papillary projection

(4) maximal diameter of solid component in mm

(5) irregular internal cyst walls

(6) presence of acoustic shadows

Parameter

Finding

Points

age of the patient in years

age in years

<age>

ascites

no

0

 

yes

1

blood flow within a papillary projection

no

0

 

yes

1

maximal diameter of the solid component in mm

diameter <= 50 mm

<diameter>

 

diameter > 50 mm

50

irregular internal cyst walls

no

0

 

yes

1

presence of acoustic shadows

no

0

 

yes

1

 

Y =

= (0.0354 * (points for age)) + (1.6159* (points for ascites)) + (1.1768 * (points for blood flow)) + (0.0697 * (points for maximum diameter of solid component)) + (0.9586 * (points for irregular internal cyst walls)) - (2.9486 * (points for acoustic shadows)) - 5.3718

 

probability of malignancy =

= 1 / (1 + EXP((-1) * Y))

 

Performance (Table 7, page 8800):

• The sensitivity is 90% and specificity is 71% at a cutoff of 10%.

• The area under the ROC curve is 0.916

 


To read more or access our algorithms and calculators, please log in or register.