Objective To develop and internally validate a deep-learning algorithm from fetal ultrasound images for the diagnosis of cystic hygromas in the first trimester. Methods All first trimester ultrasound scans with a diagnosis of a cystic hygroma between 11 and 14 weeks gestation at our tertiary care centre in Ontario, Canada were studied. Ultrasound scans with normal nuchal translucency were used as controls. The dataset was partitioned with 75% of images used for model training and 25% used for model validation. Images were analyzed using a DenseNet model and the accuracy of the trained model to correctly identify cases of cystic hygroma was assessed by calculating sensitivity, specificity, and the area under the receiver-operating characteristic (ROC) curve. Gradient class activation heat maps (Grad-CAM) were generated to assess model interpretability. Results The dataset included 289 sagittal fetal ultrasound images;129 cystic hygroma cases and 160 normal NT controls. Overall model accuracy was 93% (95% CI: 88–98%), sensitivity 92% (95% CI: 79–100%), specificity 94% (95% CI: 91–96%), and the area under the ROC curve 0.94 (95% CI: 0.89–1.0). Grad-CAM heat maps demonstrated that the model predictions were driven primarily by the fetal posterior cervical area. Conclusions Our findings demonstrate that deep-learning algorithms can achieve high accuracy in diagnostic interpretation of cystic hygroma in the first trimester, validated against expert clinical assessment.

No abstract

OBJECTIVE: To determine pregnancy ontogeny of AChE and its regulator miR-132 expression in maternal serum and fetal brain under basal and inflammatory conditions and to study the effect of Magnesium (Mg), a known neuroprotective agent, on AChE and miR-132 expression in the fetal brain. STUDY DESIGN: Pregnant rats at 16, 18 and 20 days of gestation (60 total: 8 groups, n¼5) received injections of i.p. lipopolysaccharide (LPS; 500 ug/kg) or saline (SAL) at time 0. Dams were randomized to s.c. saline or Mg (270 mg/kg loading followed by 27 mg/kg q20 min) for 2 hours prior to and 2 hours following LPS /saline injections. Rats were sacrificed 4 hours following LPS/SAL injection. Fetal brains were harvested from the treatment groups (LPS/SAL, LPS/Mg, SAL/Mg, SAL/SAL) for AChE and miR-132 mRNA expression (real-time RT-PCR) and maternal serum analyzed for AChE levels. RESULTS: Maternal serum AChE increased from e16 to e18 and e20 (169+25, 232+51,263+59 mg/ml, p< 0.05, respectively). Fetal brain AChE and miR-132 mRNA expression increased (p< 0.05) from e16 to e18 and then decreased at e20 (p< 0.05) (AChE 1.0AE0.01, 1.21AE.0.02, 0.78AE0.16; miR-132 1.0AE0.01, 1.19AE0.03, 0.86AE0.1, respectively). At e16 maternal LPS decreased fetal brain AChE (0.62AE0.1), and miR-132 mRNA expression (0.72AE0.11) compared to control (p< 0.05). Maternal Mg alone increased both fetal brain AChE and miR-132 mRNA expression at e16 (AChE1.7AE0.15; miR-132 1.7AE0.16) compared to both control and LPS fetal brains (p< 0.05). Mg treatment to LPS dams (LPS/Mg) increased fetal brain AChE and miR-132 levels (1.15AE0.04;1.25AE0.1 respectively, p< 0.05) compared to both LPS/SAL and SAL/SAL groups. The same pattern was demonstrated at e18 and e20. No significant change was demonstrated in maternal serum AChE levels following LPS or Mg compared to control. CONCLUSION: These findings suggest that fetal brain AChE activity is not regulated by miR-132. Mg treatment increases fetal brain AChE mRNA levels under basal conditions and in response to maternal LPS, suggesting that reduced ACh may increase fetal brain inflammatory responses.

No abstract