Tomoya Kobayashi

This study aimed to develop a fully automated semantic placenta segmentation model that integrates the U-Net and SegNeXt architectures through ensemble learning. A total of 218 pregnant women with suspected placental abnormalities who underwent magnetic resonance imaging (MRI) were enrolled, yielding 1090 annotated images for developing a deep learning model for placental segmentation. The images were standardized and divided into training and test sets. The performance of Placental Segmentation Network (PlaNet-S), which integrates U-Net and SegNeXt within an ensemble framework, was assessed using Intersection over Union (IoU) and counting connected components (CCC) against the U-Net, U-Net + + , and DS-transUNet. PlaNet-S had significantly higher IoU (0.78, SD = 0.10) than that of U-Net (0.73, SD = 0.13) ( p < 0.005) and DS-transUNet (0.64, SD = 0.16) ( p < 0.005), while the difference with U-Net + + (0.77, SD = 0.12) was not statistically significant. The CCC for PlaNet-S was significantly higher than that for U-Net ( p < 0.005), U-Net + + ( p < 0.005), and DS-transUNet ( p < 0.005), matching the ground truth in 86.0%, 56.7%, 67.9%, and 20.9% of the cases, respectively. PlaNet-S achieved higher IoU than U-Net and DS-transUNet, and comparable IoU to U-Net + + . Moreover, PlaNet-S significantly outperformed all three models in CCC, indicating better agreement with the ground truth. This model addresses the challenges of time-consuming physician-assisted manual segmentation and offers the potential for diverse applications in placental imaging analyses.

Signal intensity (SI) and image contrast on postmortem magnetic resonance (MR) imaging are different from those of imaging of living bodies. We sought to suppress the SI of cerebrospinal fluid (CSF) sufficiently for fluid-attenuated inversion recovery (FLAIR) sequence in postmortem MR (PMMR) imaging by optimizing inversion time (TI). We subject 28 deceased patients to PMMR imaging 3 to 113 hours after confirmation of death (mean, 27.4 hrs.). PMMR imaging was performed at 1.5 tesla, and T1 values of CSF were measured with maps of relaxation time. Rectal temperatures (RT) measured immediately after PMMR imaging ranged from 6 to 32°C (mean, 15.4°C). We analyzed the relationship between T1 and RT statistically using Pearson's correlation coefficient. We obtained FLAIR images from one cadaver using both a TI routinely used for living bodies and an optimized TI calculated from the RT. T1 values of CSF ranged from 2159 to 4063 ms (mean 2962.4), and there was a significantly positive correlation between T1 and RT (r = 0.96, P < 0.0001). The regression expression for the relationship was T1 = 74.4 * RT + 1813 for a magnetic field strength of 1.5T. The SI of CSF was effectively suppressed with the optimized TI (0.693 * T1), namely, TI = 0.693 * (77.4 * RT + 1813). Use of the TI calculated from the linear regression of the T1 and RT optimizes the FLAIR sequence of PMMR imaging.