Multi-scale and multi-view network for lung tumor segmentation.

Lung tumor segmentation in medical imaging is a critical step in the diagnosis and treatment planning for lung cancer. Accurate segmentation, however, is challenging due to the variability in tumor size, shape, and contrast against surrounding tissues. In this work, we present MSMV-Net, a novel deep learning architecture that integrates multi-scale multi-view (MSMV) learning modules and multi-scale uncertainty-based deep supervision (MUDS) for enhanced segmentation of lung tumors in computed tomography images. MSMV-Net capitalizes on the strengths of multi-view analysis and multi-scale feature extraction to address the limitations posed by small 3D lung tumors. The results indicate that MSMV-Net achieves state-of-the-art performance in lung tumor segmentation, recording a global Dice score of 55.60% on the LUNA dataset and 59.94% on the MSD dataset. Ablation studies conducted on the MSD dataset further validate that our method enhances segmentation accuracy.

SRRM2 phase separation drives assembly of nuclear speckle subcompartments.

Nuclear speckles (NSs) are nuclear biomolecular condensates that are postulated to form by macromolecular phase separation, although the detailed underlying forces driving NS formation remain elusive. SRRM2 and SON are 2 non-redundant scaffold proteins for NSs. How each individual protein governs assembly of the NS protein network and the functional relationship between SRRM2 and SON are largely unknown. Here, we uncover immiscible multiphases of SRRM2 and SON within NSs. SRRM2 and SON are functionally independent, specifically regulating alternative splicing of subsets of mRNA targets, respectively. We further show that SRRM2 forms multicomponent liquid phases in cells to drive NS subcompartmentalization, which is reliant on homotypic interaction and heterotypic non-selective protein-RNA complex coacervation-driven phase separation. SRRM2 serine/arginine-rich (RS) domains form higher-order oligomers and can be replaced by oligomerizable synthetic modules. The serine residues within the RS domains, however, play an irreplaceable role in fine-tuning the liquidity of NSs.

A study of waterside microenvironmental factors and their effects on summer outdoor thermal comfort in a Cfa-climate campus.

Outdoor thermal comfort (OTC) studies explore outdoor subjects' responses to their thermal environment, usually evaluated using the neutral temperature (NT). This study investigated the influences of microenvironmental factors around a waterbody on thermal perceptions, using questionnaires and meteorological measurements at the Central Lake of Southwest University of Science and Technology (SWUST) in Mianyang. Microenvironmental factors included sky view factor (SVF) and distance from the lake (DFL). It was found that people felt most comfortable in the shade of trees although some volunteers voted artificial canopy as their preferred thermal adaptation element. In addition, a linear regression yielded an NT of 28.44 °C in Mianyang during the summer of 2022. There were NT variations among different measurement sites (e.g., on the east shore, it was 28.18 °C on the waterside, 27.11 °C away from the lake, and 25.53 °C far from the lake; while it was 27.57 °C under the tree crown, 25.11 °C on the lawn, and 29.13 °C in the square). This variation may be due to human adaptation towards microenvironmental factors and their effects on microclimate. The variation in thermal responses owing to microenvironmental differences (different NTs at various types of sites) might be a novel finding in the field of OTC. This study provides important directions for microenvironment design in the future for OTC improvement.