Multilayer perceptron and support vector regression models for feline parturition date prediction

A crucial challenge in feline obstetric care is the accurate prediction of the parturition date during late pregnancy. The classic simple linear regression (SLR) model, which employed the fetal biparietal diameter (BPD) as the single input feature, was frequently applied for such prediction with limited accuracy. Since Multilayer Perceptron (MLP) and Support Vector Regression (SVR) are now two of the most potent scientific regression models, this study, for the first time, introduced such models as the new promising tools for feline parturition date prediction. The following features were candidate inputs for our models: biparietal diameter (BPD), litter size, and maternal weight. We observed and compared the performance results for each model. As the best-performed model, MLP delivered the highest coefficient score (0.972 ± 0.006), lowest mean absolute error score (1.110 ± 0.060), and lowest mean squared error score (1.540 ± 0.141), respectively. For the first time in this study, BPD, litter size, and maternal weight were considered the essential features for the innovative MLP and SVR modeling. With the optimized model parameters and the described analytical platform, further verification of these advanced models in feline obstetric practices is feasible.

Fetal head prior to delivery: precise measurement, using a mathematically extrapolated frontooccipital diameter

PURPOSE

To find a method that could improve the accuracy of fHC evaluation at term.

MATERIAL AND METHODS

This retrospective cohort study was based on data from the University Hospital of Zürich (USZ). Study population included singleton fetuses and newborns at term between 2015 and 2017. Fetal biometry was performed within 1 week prior to delivery. Study data consisted of two cohorts with 200 ultrasound measurements each: 200 performed by an expert, 200 by residents. fHC were compared with the newborn's head circumference (nHC). fHC was estimated using two methods: (1) fHC based on sonographically estimated frontooccipital diameter (FOD) and biparietal diameter (BPD). (2) Expected FOD was calculated as a fixed mean FOD/BPD relation, derived from biometry standards as the 50th percentile. If BPD was < 99 mm, FOD was calculated according to the formula BPD × 1.268557, If BPD was ≥ 99 mm, FOD was calculated according to the formula BPD × 1.20641443.

RESULTS

fHC was underestimated compared with nHC in the expert group: percentage error (PE =  - 3.68%, SD = 2.79), as well as in the group of residents (PE =  - 3.78%, SD = 3.20) using method 1. fHC measurement was significantly more accurate using method 2. In the expert group PE =  - 1.17%, SD = 3.08, in the group of residents PE =  - 0.95%, SD = 3.33. Bland-Altman analysis showed limits of agreement (LOA) between - 2.41338 and 0.5537828 cm for fHC.

CONCLUSIONS

Accuracy of fHC estimation could be improved by extrapolation of FOD when real-time FOD cannot be measured.

Quantitative and Qualitative Analysis of Emergency Physician Performed Biparietal Diameter Estimate for Gestational Age

BACKGROUND

An accurate estimation of fetal gestational age is essential for the management of pregnant patients who present to the emergency department (ED). Point-of-care-ultrasound (POCUS) is an integral part of emergency medicine training and includes measurement of fetal gestational age by the biparietal diameter (BPD) method.

OBJECTIVES

In this study we performed a quantitative assessment of emergency physician (EP)-performed BPD estimate of gestational age to identify the percentage of studies performed in our department that had an estimated gestational age within 14 days of an adjusted radiological or obstetrical estimation.

METHODS

We performed a chart review of our ED ultrasound database and electronic medical records for cases where a BPD measurement was performed by an EP. We recorded the ED gestational age estimate in days and the radiological or obstetrical estimate of gestational age in days. We then calculated the difference in days between the two examinations. We used a normal binomial approximation to calculate 95% confidence intervals. A secondary analysis looked at the quality of the images based on measurement placement and the view obtained.

RESULTS

Seventy-four cases met eligibility criteria; of those, 67 (91%) had a gestational age estimation within 14 days of the adjusted radiological or obstetrical estimate (95% confidence interval 81-96%).

CONCLUSION

This study shows that EP-performed BPD measurements for gestational age are quantitatively accurate, with 91% of estimates within 14 days of a standard radiological or obstetrical estimation.

Relationship between biparietal diameter/ventricular ratio and neurodevelopmental outcomes in non-handicapped very preterm infants

PURPOSE

Preterm infants are at high risk for brain damage and long-term neurodevelopmental problems. Cranial ultrasonography is the main neuroimaging technique for very low birth weight infants. Ventricle size and its ratio to brain volume contribute very important information about the central nervous system of preterm babies. We calculated biparietal diameter/ventricular ratio of preterm infants using cranial ultrasonography and evaluate the relationship between this ratio and neurodevelopment.

METHODS

Cranial measurements were derived using routine ultrasonographic scanning. Transverse brain length, or biparietal diameter (BPD), was considered a representation of the total brain, ventricular index (VI) and thalamo-occipital distance (TOD) length were used to represent the ventricles, and their ratio was accepted as a measure of the tissue portion of the brain. The ratio of BPD to the sum of left and right VI and TOD values was recorded as BPD/(VI+TOD) ratio.

RESULTS

Data from a total of 482 patients were analyzed. The mean gestational age was 27.6 (24-29.6) weeks and the mean birth weight was 1010 (350-1390) g. The mean BPD/(VI+TOD) ratio was 32.90 (± 2.32). At 24 months corrected age, the patients' mean MDI score was 78.64 (± 13.29) and mean PDI score was 79.49 (± 14.31). When patients with and without NDI were compared, there were significant differences between the groups in terms of BPD/(VI+TOD) ratio, MDI, and PDI (p < 0.001, p < 0.001, p < 0.001, respectively).

CONCLUSION

The BPD/ventricle ratio can be calculated using two-dimensional measurements in VLBW infants and reduced BPD/ventricle ratio was associated with poor neurodevelopmental outcomes.

TRIAL REGISTRATION

NCT02848755.

Obstetric forceps dimensions and the newborn head biometry: Time for an update

Obstetric forceps were invented in the 1600s to assist vaginal delivery of term babies following prolonged labour. This probably explains their design, with a narrow interblade distance and long blade length, to fit a severely moulded fetal head. However, in modern obstetric practice protracted labour is avoided, yet our research has shown that over 400 years forceps dimensions have remained largely unchanged. We believe it is time to optimise these dimensions based on biometry of the term, newborn baby's head, with the head width (biparietal diameter) and head length (mentovertical diameter) correlating with interblade distance and blade length respectively. We hypothesise that doing so should reduce the incidence of neonatal complications associated with forceps assisted delivery and it is also possible that the amended shape might be associated with better outcomes for women. In this article we present our rationale for the optimisation of the forceps dimensions based on the findings of our previous systematic review and an original series of mentovertical and biparietal diameter measurements using laser scanning technology.

Application of the LMS method of constructing fetal reference charts: comparison with the original method

Objectives: In view of the concern expressed about the current references, new references for fetal biparietal diameter and head circumference should be constructed for contemporary local populations.Methods: We conducted a retrospective cross-sectional study in two hospitals in Guangdong, Southern China. Fetal biparietal diameter and head circumference percentiles regression were fitted using Cole's LMS method. The BPD and HC data were then transformed into Z-scores that were calculated using two series of reference equations obtained from two methods: Cole's LMS method and the original "mean and SD method." Each Z-score distribution was presented as the mean and standard deviation. Finally, the sensitivity and specificity of each reference for identifying fetuses <2.5th or >97.5th percentile (based on the observed distribution of Z-scores) were calculated. The misclassified number and Youden's index were listed.Results: A total of 17,974 biparietal diameter and 18,269 head circumference measurements were chosen to establish a reference chart. The LMS method could fit the local population better than the "mean and SD method" as it had a lower number of misclassified fetuses and a higher Youden's index.Conclusion: The Cole's LMS method was able to construct a satisfied reference range of fetal head sizes in Southern China.

Is the biparietal diameter of fetuses in late gestation too variable to predict readiness for cesarean section in dogs?

Correct assessment of readiness for cesarean section is essential for timing elective cesarean section during late pregnancy in the bitch. In humans, biparietal diameter is sufficiently precise and accurate and used in a clinical setting daily. The objectives of this study were to determine whether fetal biparietal diameter in late gestation in the dog could be used to predict readiness for cesarean section by having reached a minimum cut-off value and to correlate the biparietal diameter to birth weight. The biparietal diameter of 208 puppies in 34 litters from 31 English bulldog bitches and 660 puppies in 78 litters from 70 Boerboel bitches were measured immediately after delivery by cesarean section, performed at full term, using digital calipers. At the same time the birth weight of the same 208 English bulldog puppies and 494 of the same Boerboel puppies in 59 litters from 54 bitches was measured by means of an electronic scale. With a cesarean section, all the puppies in a litter are delivered simultaneously and readiness for cesarean section must be determined for a litter. The minimum, median and maximum biparietal diameter varied from 21.1 to 47.8, 32.9 to 50.0 and 34.2-58.2 mm, respectively, among English bulldog litters and from 18.4 to 48.7, 35.5 to 49.7 and 39.8-54.3 mm among Boerboel litters. This large variation suggests that biparietal diameter is too variable within and among litters to be useful as a means of determining readiness for cesarean section.

Fetal size and growth velocity in chronic hypertension

OBJECTIVE

To investigate longitudinal fetal growth and growth velocity for commonly measured biometric parameters in women with chronic hypertension.

METHODS

Two centre retrospective European study of women with chronic hypertension ascertained at pregnancy booking. Ultrasound measurements of head circumference (HC), abdominal circumference (AC) and femur length (FL) were used to derive longitudinal fetal growth charts derived using functional linear discriminant analysis (FLDA). These were compared to existing cross sectional and longitudinal charts, as was birthweight.

RESULTS

282 women with a median of 3 third trimester ultrasound examinations were included. Gestation at delivery was 37.5weeks (SD 2.68), birthweight 3049g (SD 785). Birthweight <10th percentile found in 15.6% deliveries, >90th percentile 20.2%. Fetal size curves derived from women with chronic hypertension were no different to cross sectional and longitudinal charts for a normal population. Compared to a standard longitudinal biometry chart, growth velocity (mm/day) in chronic hypertension was higher for AC and FL at 30-32weeks (AC 1.447vs 1.357 p<0.05; FL 0.296vs 0.269 p<0.01) and 34-36weeks (AC 1.325vs 1.140 p<0.01; FL 0.248vs 0.198 p<0.01).

CONCLUSIONS

In women with chronic hypertension there is an excess of both SGA and LGA babies compared to population standards. Growth velocity of the AC and FL was greater after 30weeks compared to a normal population.

Characterization of fetal growth by repeated ultrasound measurements in the wild guinea pig (Cavia aperea)

Fetal growth during pregnancy has previously been studied in the domesticated guinea pig (Cavia aperea f. porcellus) after dissecting pregnant females, but there are no studies describing the fetal growth in their wild progenitor, the wild guinea pig (C aperea). In this study, 50 pregnancies of wild guinea pig sows were investigated using modern ultrasound technique. The two most common fetal growth parameters (biparietal diameter [BPD] and crown-rump-length [CRL]) and uterine position were measured. Data revealed similar fetal growth patterns in the wild guinea pig and domesticated guinea pig in the investigated gestation period, although they differ in reproductive milestones such as gestation length (average duration of pregnancy 68 days), average birth weight, and litter mass. In this study, pregnancy lasted on average 60.2 days with a variance of less than a day (0.96 days). The measured fetal growth parameters are strongly correlated with each (R = 0.91; P < 0.001) other and with gestational age (BPD regression equation y = 0.04x - 0.29; P < 0.001 and CRL regression equation y = 0.17x - 2.21; P < 0.01). Furthermore, fetuses in the most frequent uterine positions did not differ in their growth parameters and were not influenced by the mother ID. Our results imply that ultrasound measurement of a single fetal growth parameter is sufficient to reliably estimate gestational age in the wild guinea pig.