Search for: All records

Abstract Image segmentation of the liver is an important step in treatment planning for liver cancer. However, manual segmentation at a large scale is not practical, leading to increasing reliance on deep learning models to automatically segment the liver. This manuscript develops a generalizable deep learning model to segment the liver on T1-weighted MR images. In particular, three distinct deep learning architectures (nnUNet, PocketNet, Swin UNETR) were considered using data gathered from six geographically different institutions. A total of 819 T1-weighted MR images were gathered from both public and internal sources. Our experiments compared each architecture’s testing performance when trained both intra-institutionally and inter-institutionally. Models trained using nnUNet and its PocketNet variant achieved mean Dice-Sorensen similarity coefficients>0.9 on both intra- and inter-institutional test set data. The performance of these models suggests that nnUNet and PocketNet liver segmentation models trained on a large and diverse collection of T1-weighted MR images would on average achieve good intra-institutional segmentation performance. 

Abstract The ability of organisms to cross ecosystem boundaries is an important catalyst of evolutionary diversification. The genus Poecilia (mollies and guppies) is an excellent system for studying ecosystem transitions because species display a range of salinity and dietary preferences, with herbivory concentrated in the subgenus Mollienesia. We reconstructed ancestral habitats and diets across a phylogeny of the genus Poecilia, evaluated diversification rates and used phylogenetically independent contrasts to determine whether diet evolved in response to habitat transition in this group. The results suggest that ancestors of subgenus Mollienesia were exclusively herbivorous, whereas ancestral diets of other Poecilia included animals. We found that transitions across euryhaline boundaries occurred at least once in this group, probably after the divergence of the subgenus Mollienesia. Furthermore, increased salinity affiliation explained 24% of the decrease in animals in the gut, and jaw morphology was associated with the percentage of animals in the gut, but not with the percentage of species occupying saline habitats. These findings suggest that in the genus Poecilia, herbivory evolved in association with transitions from fresh to euryhaline habitats, and jaw morphology evolved in response to the appearance of herbivory. These results provide a rare example of increased diet diversification associated with the transition from freshwater to euryhaline habitats.