Premium
P07.01: Artificial intelligence (AI) weights the importance of clinical and sonographic factors predicting malignancy in unilocular‐solid cysts before surgery
Author(s) -
Chiappa V.,
Bogani G.,
Fruscio R.,
Franchi D.,
Urbinati A.M. Vidal,
Giuliani D.,
Marchette M. Delle,
Brunetti F.,
Murru G.,
Raspagliesi F.
Publication year - 2019
Publication title -
ultrasound in obstetrics and gynecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.202
H-Index - 141
eISSN - 1469-0705
pISSN - 0960-7692
DOI - 10.1002/uog.20933
Subject(s) - medicine , malignancy , referral , breast cancer , weighting , radiology , cancer , family medicine
The primary outcome was SGA, defined as a baby with birth weight < 10th centile according with INTERGROWTH-21st standards. Sensitivity and specificity were calculated. Results: UtA Doppler assessment and birth weight were available in 864 women. Amongst those women, neonatal SGA occurred in 51 (5.90%), UtA was abnormal in 91% of all cases. The overall sensitivity, specificity, positive and negative predictive value were, 33, 88, 16, and 95%, respectively. In the low and intermediate-low risk group sensitivity and negative predictive value were similar to the all population (33, and 96%, respectively) (group A and B). SGA rate detection in the intermediate-low risk group doubled (from 8 to 16 cases). In the intermediate-high and high risk groups (C, D), sensitivity was higher (58%); with a lower negative predictive value (91%). Conclusions: UtA Doppler assessment in the second trimester, as a part of a multiparamenter test for screening for SGA, improves detection rate, especially in the subgroup of patients with risk factors for SGA. Negative predictive value in low-risk women can justify its use routinely.