PubMed Health⌕ Search

Biomedical subjects

N Vlassis

Publications and source records attributed to N Vlassis.

2 recordsLinked to original sources
PubMed2003

Fetal loss following ultrasound diagnosis of a live fetus at 6-10 weeks of gestation.

OBJECTIVE: To examine prospectively the value of demographic characteristics and ultrasound findings in the prediction of subsequent fetal loss in pregnancies with live fetuses at 6-10 weeks of gestation. METHODS: Transvaginal ultrasound examination was performed in 866 pregnancies at 6-10 weeks of gestation. The relation of demographic data and ultrasound findings at the time of the initial assessment to subsequent fetal loss was examined. RESULTS: In the 668 singleton pregnancies with live fetuses and complete follow-up there were 50 (7.5%) fetal losses. The incidence of fetal loss increased significantly with maternal age and decreased with gestation. In the pregnancies resulting in fetal loss, compared to those with live births, the incidence of vaginal bleeding and cigarette smoking was higher, the fetal heart rate was significantly lower and the gestation sac diameter was smaller but the yolk sac diameter was not significantly different. CONCLUSION: In pregnancies with a live fetus at 6-10 weeks' gestation the rate of subsequent fetal loss is related to maternal age, gestation, cigarette smoking, history of vaginal bleeding and the ultrasound findings of small gestation sac diameter and fetal bradycardia, relative to crown-rump length.

Female↗
PubMed2003

Efficient greedy learning of gaussian mixture models.

This article concerns the greedy learning of gaussian mixtures. In the greedy approach, mixture components are inserted into the mixture one after the other. We propose a heuristic for searching for the optimal component to insert. In a randomized manner, a set of candidate new components is generated. For each of these candidates, we find the locally optimal new component and insert it into the existing mixture. The resulting algorithm resolves the sensitivity to initialization of state-of-the-art methods, like expectation maximization, and has running time linear in the number of data points and quadratic in the (final) number of mixture components. Due to its greedy nature, the algorithm can be particularly useful when the optimal number of mixture components is unknown. Experimental results comparing the proposed algorithm to other methods on density estimation and texture segmentation are provided.

Algorithms↗