Deep learning to estimate gestational age from blind ultrasound sweeps of the gravid abdomen

Background: Ultrasound is indispensable to gestational age estimation, and thus to quality obstetric care, yet high equipment cost and need for trained sonographers limit its use in low-resource settings. Methods: From September 2018 through June 2021, we recruited 4,695 pregnant volunteers in North Carolina and Zambia and obtained blind ultrasound sweeps (cineloops) of the gravid abdomen alongside standard fetal biometry. We trained a neural network to estimate gestational age from the sweeps and, in three test sets, assessed performance of the model and biometry against previously established gestational age. Results: In our main test set, model mean absolute error (MAE) was 3.9 days (standard error [SE] 0.12) vs. 4.7 days (SE 0.15) for biometry (difference -0.8 days; 95% CI -1.1, -0.5; p<0.001). Results were similar in North Carolina (difference -0.6 days, 95% CI -0.9, -0.2) and Zambia (-1.0 days, 95% CI -1.5, -0.5). Findings were supported in the test set of women who conceived by in vitro fertilization (model MAE 2.8 days [SE 0.28] vs. 3.6 days [SE 0.53] for biometry; difference -0.8 days, 95% CI -1.7, 0.2), and in the set of women from whom sweeps were collected by untrained users with low-cost, battery-powered devices (model MAE 4.9 days [SE 0.29] vs. 5.4 days [SE 0.28] for biometry; difference -0.6, 95% CI -1.3, 0.1). Conclusions: Our model estimated gestational age more accurately from blindly obtained ultrasound sweeps than did trained sonographers performing fetal biometry. These results presage a future where all pregnant people - not just those in rich countries - can access the diagnostic benefits of sonography.

Can trans-cerebellar diameter supersede other fetal biometry in measuring gestational age? A prospective study

Background Estimation of gestational age plays a pivotal role in day to day clinical practice for appropriate management of newborn. The trans-cerebellar diameter can predict gestational age in cases of variations of fetal head shape such as dolichocephaly and brachycephaly or even when fetus is in posterior position. Trans-cerebellar diameter (TCD) may be useful in gestational age estimation. Aim and methodology To determine the accuracy of trans-cerebellar diameter in the estimation of gestational age of fetus and comparing the conventional parameters like head circumference, biparietal diameter, abdominal circumference and femur length in the measurement of gestational age. This is a prospective, cross-sectional analytical study done in the Radiology Department in those patients who were referred to the department for a routine antenatal ultra-sonogram checkup. Results Comparison of trans-cerebellar diameter with gestational age derived from last menstrual period indicates that there is a linear relationship throughout the gestational ages. There is a strong correlation of trans-cerebellar diameter with other conventional parameters namely biparietal diameter, femur length and abdominal circumference of which Femur length correlates well with the trans-cerebellar diameter. Correlation of transcerebellar diameter with gestational age shows a R2 of 0.995 (p value < 0.001). Conclusion Trans-cerebellar diameter is best in estimation of gestational age when compared to biparietal diameter, femur length and abdominal circumference. The regression formula derived from trans-cerebellar diameter measurement can be applied to determine the gestational age of fetus.

INNOVATE GESTATIONAL AGE ESTIMATION MODEL FOR IRAQI FETUSES BASED ON ULTRASOUND IMAGES MEASUREMENTS

Imaging by Ultrasound (US) is an accurate and useful modality for the assessment of gestational age (GA), estimation fetal weight, and monitoring the fetal growth during pregnancy, is a routine part of prenatal care, and that can greatly impact obstetric management. Estimation of GA is important in obstetric care, making appropriate management decisions requires accurate appraisal of GA. Accurate GA estimation may assist obstetricians in appropriately counseling women who are at risk of a preterm delivery about likely neonatal outcomes, and it is essential in the evaluation of the fetal growth and detection of intrauterine growth restriction. There are many formulas are used to estimate fetal GA in the world, but it's not specify for Iraqi population and leading to some error in GA estimation results, so the objective of this study is to innovate GA estimation model for Iraqi people. This study was performed in the department of Obstetrics and Gynecology in Al- Yarmouk Teaching Hospital and AL- Alawiya Teaching Hospital in Baghdad, Iraq, during 2019 on 200 pregnant women of singleton and normal pregnancies, fetal GA (20-40) weeks (W). The obtained dataset (fetal biometry), were utilized to create GA estimation model in Iraq using IBM SPSS Version 23 software package (IBM^® Software). The statistical analysis of proposed GA model showed, the correlation (R) of model is 0.987 it is very high value and this is a good result to obtain the best regression model. as well as the Std error of Estimation was 0.61095 this is very small value and indicate the best result. The significant of model P=0.000 That means the model, as a whole, is a significant fit to the data (because P < 0.05). Keywords: Ultrasound, Fetal Biometry, Estimation Gestational Age, SPSS.