A Comparative Analysis of U-Net and Vision Transformer Architectures in Semi-Supervised Prostate Zonal Segmentation.

The precise segmentation of different regions of the prostate is crucial in the diagnosis and treatment of prostate-related diseases. However, the scarcity of labeled prostate data poses a challenge for the accurate segmentation of its different regions. We perform the segmentation of different regions of the prostate using U-Net- and Vision Transformer (ViT)-based architectures. We use five semi-supervised learning methods, including entropy minimization, cross pseudo-supervision, mean teacher, uncertainty-aware mean teacher (UAMT), and interpolation consistency training (ICT) to compare the results with the state-of-the-art prostate semi-supervised segmentation network uncertainty-aware temporal self-learning (UATS). The UAMT method improves the prostate segmentation accuracy and provides stable prostate region segmentation results. ICT plays a more stable role in the prostate region segmentation results, which provides strong support for the medical image segmentation task, and demonstrates the robustness of U-Net for medical image segmentation. UATS is still more applicable to the U-Net backbone and has a very significant effect on a positive prediction rate. However, the performance of ViT in combination with semi-supervision still requires further optimization. This comparative analysis applies various semi-supervised learning methods to prostate zonal segmentation. It guides future prostate segmentation developments and offers insights into utilizing limited labeled data in medical imaging.

Assessing the impact of trade policy uncertainty on pollution emissions: an analysis of Chinese firms' green transformation.

Oscillations in the global trade milieu have exacerbated the ambiguity experienced by Chinese enterprises, thereby influencing their ecological transition. The ongoing debate over whether trade uncertainty augments corporate emissions, exacerbating pollution, or attenuates emissions, encouraging sustainable production, has yet to reach a consensus. The current investigation employs a textual analysis methodology to explore the influence of trade policy uncertainty on pollution emissions, by sourcing indicators of trade policy uncertainty that echo firm-level uncertainty within the period 2008 to 2021. Utilizing the fixed effects model for our analysis, the findings substantiate that escalated uncertainty at the micro-level catalyzes an increase in pollution emissions originating from firms. Crucially, we find that risk diversification and innovation bolster firms' capacities to manage pollution under escalating uncertainty. Furthermore, our estimation reveals that enterprises with low market competitiveness, high financial constraints, and moderate overseas market share are most significantly impacted, whereas those with robust patent portfolios remain largely unaffected. This study carries considerable implications for firms striving to achieve an ecological transition and offers insights for fostering sustainable and high-quality global economic development.

Simultaneous Optimization of Donor/Acceptor Pairs and Device Specifications for Nonfullerene Organic Solar Cells Using a QSPR Model with Morphological Descriptors.

Optimally efficient organic solar cells require not only a careful choice of new donor (D) and/or acceptor (A) molecules but also the fine-tuning of experimental fabrication conditions for organic solar cells (OSCs). Herein, a new framework for simultaneously optimizing D/A molecule pairs and device specifications of OSCs is proposed, through a quantitative structure-property relationship (QSPR) model built by machine learning. Combining the device bulk properties with structural and electronic properties, the built QSPR model achieved unprecedentedly high accuracy and consistency. Additionally, a large chemical space of 1 942 785 D/A pairs is explored to find potential synergistic ones. Favorable device bulk properties such as the root-mean-square of surfaces roughness for D/A blends and the D/A weight ratio are further screened by grid search methods. Overall, this study indicates that the simultaneous optimization of D/A molecule pairs and device specifications by theoretical calculations can accelerate the improvement of OSC efficiencies.