In recent years, there has been a growing demand to stylize a given 3D scene to align with the artistic style of reference images for creative purposes. While 3D Gaussian Splatting (GS) has emerged as a promising and efficient method for realistic 3D scene modeling, there remains a challenge in adapting it to stylize 3D GS to match with multiple styles through automatic local style transfer or manual designation, while maintaining memory efficiency for stylization training. In this paper, we introduce a novel 3D GS stylization solution termed Multi-StyleGS to tackle these challenges. In particular, we employ a bipartite matching mechanism to automatically identify correspondences between the style images and the local regions of the rendered images. To facilitate local style transfer, we introduce a novel semantic style loss function that employs a segmentation network to apply distinct styles to various objects of the scene and propose a local-global feature matching to enhance the multi-view consistency. Furthermore, this technique can achieve memory-efficient training, more texture details and better color match. To better assign a robust semantic label to each Gaussian, we propose several techniques to regularize the segmentation network. As demonstrated by our comprehensive experiments, our approach outperforms existing ones in producing plausible stylization results and offering flexible editing.

Lymph node metastasis is an important decision-making basis for lung cancer staging and clinical treatment, and will also significantly affect the prognosis of patients. At present, there is no research to identify lymph node metastasis in non-small cell lung cancer (NSCLC) using cell-cell interaction network (CCIN). We believe that utilizing the characteristics of CCIN can not only distinguish whether different patients have lymph node metastasis, but also help reveal the development process of lung cancer from the perspective of molecular interaction networks. This study utilized graph embedding technology and Fisher's exact test to analyze CCIN, and then combined with a classifier for identifying subtypes and lymph node metastasis of NSCLC. The experimental results showed that the performance of our method is similar to that of image-based methods, and the CCINs at each stage of adenocarcinoma of the lung (LUAD) and lung squamous cell carcinoma (LUSC) were analyzed to identify the differential biological pathways, and further characterized the lymph node metastasis development profile with significant interaction pairs. It is worth noting that the interaction pair features identified by the our method highly were enriched in some known pathways related to the development of lung cancer, thus proving the effectiveness of our method.

The rapid development of single-cell technology has opened up a whole new perspective for identifying cell types in multicellular organisms and understanding the relationships between them. Distinguishing different cell types and subtypes can identify the components of different immune cells and different tumor clones in the tumor microenvironment, which is the basic work of tumor heterogeneity analysis and can help researchers understand the mechanism of tumor immune escape. Existing algorithms treat both cell types and subtypes as populations of cells with specific gene expression patterns, which is not conducive to accurate cell typing. For that, we proposed a cell similarity metric that unifies cell type recognition and subtype identification (UCRSI), with the assumption that selectively expressed genes represent differences in underlying cell type with on/off manner, while differences in expression level represent different cell subtype with more/less manner. Our method calculates these two kinds of differences separately, and then combines them using a consensus adjacency matrix, and finally cell typing is completed using spectral clustering algorithm. The results show that UCRSI can reconstruct expert annotation of single-cell RNA sequencing datasets more robustly than existing methods. And, UCRSI is useful for analyzing tumor heterogeneity and improving visualization of large-scale cell clustering.

Perception and reasoning are two representative abilities of intelligence, and Abductive Learning (ABL) is targeted at unifying the two abilities in a mutually beneficial way, where the neural model learns to perceive the primitive logic facts from raw data, and logical reasoning can exploit domain knowledge to correct the wrongly perceived facts for training the neural model. However, there are three flaws in ABL: (1) While logical reasoning corrects the wrongly perceived facts, the corrected facts may contain redundancy. (2) Some complex facts are hard to correct, therefore no supervision can be provided by the corresponding samples in dataset. (3) The correction of logical reasoning induces a large number of spurious facts, which degrades the performance of the model. In this paper, we focus on these issues and argue that the lack of overall exploitation of domain knowledge is the root cause. We propose a Smarter Abductive Learning (S-ABL) framework. In this framework, we first sum up some extra knowledge and propose three mechanisms via extra knowledge, namely simplification, filtration, augmentation. We exploit the extra knowledge to simplify, filter and augment corrected facts to improve the quality and quantity of the corrected facts. The effectiveness of our framework has been verified in the math word problem. We experiment on dataset Math23k and MAWPS. In weak supervision, our method achieves 64.8% and 52.9% accuracy, outperforming the existing baseline by 5.4% points and 7.3% points on dataset Math23k and MAWPS respectively.

Although the neuroimaging measures build a bridge between genetic variants and disease phenotypes, an assessment of single nucleotide variants changes in brain structure and their clinically influence on the progression of Alzheimer's disease remain largely preliminary. Note that each variant has very weak correlation signal to neuroimaging measures or Alzheimer's disease phenotypes. Therefore, traditional sparse regression-based image genetics approaches confront with unresolvable features, relative high regression error or inapplicability of high-dimensional data. Adopting an [Formula: see text] regularization method, we significantly elevate the regression accuracy of imaging genetics compared with group-sparse multitask regression method. With further analysis on the simulation results, we conclude that multiple regression tasks model may be unsuitable for image genetics. In addition, we carried out a whole genome association analysis between genetic variants (about 388 million loci) and phenotypes (cognition normal, mild cognitive impairment and Alzheimer's disease) with using the [Formula: see text] regularization method. After annotating the effect of all variants by Ensembl Variant Effect Predictor (VEP), our method locates 33 missense variants which can explain 40% phenotype variance. Then, we mapped each missense variant to the nearest gene and carried out pathway enrichment analysis. The Notch signaling pathway and Apoptosis pathway have been reported to be related to the formation of Alzheimer's disease.