This study introduces a modified parallel net (MPNET), a novel deep learning model designed for accurate segmentation and quantification of visceral and superficial adipose tissues. This was used to quantify the visce...
详细信息
This study introduces a modified parallel net (MPNET), a novel deep learning model designed for accurate segmentation and quantification of visceral and superficial adipose tissues. This was used to quantify the visceral and superficial adipose tissues found at the L3 levels of vertebra in CT scans. This will be used to predict the likelihood of the patient developing diabetes or cardiovascular diseases from existing CT scan data. MPNET was compared with state-of-the-art models like UNET, R2UNET, UNET++, and nnUNET. This approach advances the accuracy and efficiency of image segmentation demonstrating a faster learning curve and lower losses at early epochs than traditional models., We developed and validated using a limited dataset of 14 single-slice DICOM files for each patient extracted from the National Health Service UK. The outputs from MPNET not only matched but often exceeded traditional metrics such as the Dice coefficient and IoU in nuanced anatomical delineation, providing greater clinical realism and applicability in segmentation results. As a pilot study, this research paves the way for a forthcoming validation study on a larger and more ethnically diverse dataset.
Epicardial adipose tissue (EAT) is contiguous with arteries and myocardium. An increase in the volume of EAT may lead to adverse cardiovascular events. Therefore, quantification of EAT is necessary. The purpose of thi...
详细信息
Epicardial adipose tissue (EAT) is contiguous with arteries and myocardium. An increase in the volume of EAT may lead to adverse cardiovascular events. Therefore, quantification of EAT is necessary. The purpose of this paper is to employ a more than helpful algorithm for EAT segmentation and quantification. First, we used a simple convolutional neural network to select EAT slices, which significantly reduced oversegmentation. Then, we employed multiscale residual attention Unet (MRA-Unet) to achieve EAT segmentation based on the selected slices. Finally, we calculated the segmented volume to quantify EAT. We used 33/103 patients to test the model. The average Dice score for EAT segmentation was 0.883. For EAT quantification, the Pearson and concordance correlation coefficients reached 0.973 and 0.971, respectively. The results showed that our algorithm had strong agreement and consistency with expert. Our method performed efficient quantification and had strong consistency and agreement with the volume manually marked by experts. This algorithm can be used as a tool to assist in the clinical quantification of EAT. By combining different measurements to predict adverse cardiovascular and heart disease events, it has the potential to be applied for clinical use in the future.
暂无评论