A comparison of methods for preoperative discrimination between malignant and benign adnexal masses: The development of a new logistic regression model

Comparison of ‘pattern recognition’ and logistic regression models for discrimination between benign and malignant pelvic masses: a prospective cross validation

“…The model of Tailor and coworkers had a sensitivity of 93% and a specificity of 90%, when a risk of malignancy of > 25% was used to indicate malignancy 4 . The corresponding figures for the logistic regression model suggested by Timmerman and colleagues were 96% and 87% 5 . In the two studies cited, the mathematical models were created and tested in the same group of patients.…”

“…However, for less experienced sonographers, ultrasound methods other than pattern recognition might be preferable. Recently, multiple logistic regression models or artificial neural networks using clinical information and results of gray-scale and Doppler ultrasound examination have been advocated as excellent methods Correspondence: Dr L. Valentin, Department of Obstetrics and Gynecology, Lund University, University Hospital, Malmö, SE 205 02 Malmö, Sweden (e-mail: lil.valentin@obst.mas.lu.se) for calculating individual risks of malignancy in adnexal masses [4][5][6] . The multiple logistic regression model suggested by Tailor and coworkers 4 includes information on the woman's age, the presence of papillary projections in the mass, and the highest time-averaged maximum velocity recorded from the tumor using Doppler ultrasound.…”

“…The multiple logistic regression model suggested by Tailor and coworkers 4 includes information on the woman's age, the presence of papillary projections in the mass, and the highest time-averaged maximum velocity recorded from the tumor using Doppler ultrasound. The model of Timmerman and colleagues 5 uses information on menopausal status, the presence of papillary projections, the color content of the color Doppler tumor scan, and serum CA 125 values. The model of Tailor and coworkers had a sensitivity of 93% and a specificity of 90%, when a risk of malignancy of > 25% was used to indicate malignancy 4 .…”

Imaging in gynecology

“…In studies where pattern recognition was compared to other ultrasound methods (Lerner score [111], risk of malignancy index [112], the Tailor risk calculation model [113], the Timmerman risk calculation model [114] and the mathematical risk calculation models designed in the International Ovarian Tumor Analysis (IOTA) study [115]), pattern recognition was superior to the other methods for distinguishing benign from malignant adnexal masses (Table 7) [115][116][117].…”

“…However, whether this is indeed the case has not been studied. Some of the risk calculation models (the Timmerman model [114], the Tailor model [113]) and scoring systems (the Sassone score [118], the Lerner score [111], the De Priest score [119]) did not perform very well when they were tested prospectively [116,117,120,121].…”