Accurate Prenatal Diagnosis of Cleft Palate Using Magnetic Resonance Imaging with Slice-to-Volume Reconstruction

Abstract

Cleft palate (CP) is one of the most common human birth defects. The routine clinical technique for prenatal diagnosis of CP is ultrasound (US). However, US has many limitations on the diagnosis of CP especially posterior CP. In this work, we employed a 3D super-resolution reconstruction method to reconstruct 3D isotropic volumetric images from a series of 2D images and evaluate its feasibility to the diagnosis of CP. Quantitative comparison between original 2D and the reconstructed 3D images shows that the reconstructed 3D images have better image quality than the original 2D images in terms of SNR (21.48 ± 0.58 vs. 14.80 ± 2.59, p = 0.021), CNR (73.13 ± 10.96 vs. 42.97 ± 6.75, p = 0.022), image quality score (3.4 ± 0.7 vs. 3.0 ± 0.8, p = 0.046), and diagnostic confidence score (3.8 ± 0.4 vs. 3.1 ± 0.9, p = 0.038). Using the findings at birth as standard reference, the sensitivity (SE), specificity (SP), positive and negative predictive values (PPV and NPV), and the accuracy (ACC) of conventional 2D method and the proposed 3D method were calculated, and all participants were correctly identified by the radiologists based on the reconstructed 3D images. There was one participant with fetal cleft lip and palate that were not correctly identified based the original 2D images. Therefore, the reconstructed 3D images can improve the accuracy and efficiency of the prenatal diagnosis of CP. It can serve as a complementary tool for the diagnosis of CP.