| Objective: The purpose of this study is to evaluate the fetal lung maturity(FLM)of pregnant women with high blood pressure during pregnancy(HDP)by using artificial intelligence quantitative ultrasound texture analysis.To assess their birth for neonatal respirato ry system diseases,such as neonatal respiratory distress syndrome(NRDS),and transient tachypnea,the feasibility and clinical value,to further explore a non-invasive,accurate,safe and convenient method for FLM evaluation.For guiding the appropriate use of glucocorticoid,choose a suitable delivery time to provide a reference basis,accumulation,collecting relevant data of the HDP pregnant women and fetus in Inner Mongolia,evaluate the fetal lung development situation,for predicting neonatal respir atory system disease,the risk of a pre-term birth control and prevention,provide a scientific basis for maternal-fetal health management and so on.Materials and Methods: From January 2019 to January 2021,60 HDP pregnant women who met the inclusion and exclusion criteria in our hospital were selected.The Gevolusone10 professional gynecology and obstetrics color ultrasound diagnostic instrument was used for scanning.The probe frequency was 2-6MHz,and the heart preset conditions in the obstetrics conditions of middle and late pregnancy were selected.Harmonics were used,and smoothing,adaptive image processing,frame correlation and automatic gain control Settings were disallowed.The pregnant woman is in supine or side decubitus position.Based on the four-chamber view of the fetus,the whole fetal chest structure is shown,the fetal spine is about 3 or9 o ’clock horizontal.After the collection is completed,the image is saved without any form of compression and the format is DICOM.Will save good image upload AI quantitative ultrasound image texture analysis,get the test results showed that the fetus fetal lung maturity and comprehensive of neonatal respiratory morbidity(NRM)high or low risk.Apgar score and p H value of umbilical artery blood gas a nalysis were used as criteria to evaluate the presence of NRDS in neonates after elective delivery within 48 hours.When Apgar score < 8 and p H < 7.2,it was judged that the newborn had NRDS.If Apgar score ≥8 and p H ≥7.2,it was judged that the newborn had no NRDS.If the newborn only met 1 item,the basis for judgment was insufficient and it was temporarily excluded.Results: The actual analysis and statistics of 57 cases in this study showed that the incidence of fetal NRDS in pregnant women with HDP was 47.36%.The sensitivity,specificity,positive likelihood ratio,negative likelihood ratio,positive predictive value,negative predictive value and Youden index were 92.59%,80.00%,4.63,0.09,80.64%,92.30% and 0.725 respectively for the evaluation of FLM in pregnant women with HDP.Conclusion: the quantitative ultrasound image texture analysis based on artificial intelligence assessment of pih pregnant women FLM method has higher sensitivity and negative predictive value,can more accurately and effectively evaluate intrauterine fetal FLM,noninvasive,safety,high efficiency,and its repeatability is easy to be a doctor and accepted by the client,and evaluation for clinical intervention delivery time to provide certain scientific basis,has certain cl inical value and social value,its application prospect. |
PDF Full Text Request