RSAFormer: A method of polyp segmentation with region self-attention transformer.

Colonoscopy has attached great importance to early screening and clinical diagnosis of colon cancer. It remains a challenging task to achieve fine segmentation of polyps. However, existing State-of-the-art models still have limited segmentation ability due to the lack of clear and highly similar boundaries between normal tissue and polyps. To deal with this problem, we propose a region self-attention enhancement network (RSAFormer) with a transformer encoder to capture more robust features. Different from other excellent methods, RSAFormer uniquely employs a dual decoder structure to generate various feature maps. Contrasting with traditional methods that typically employ a single decoder, it offers more flexibility and detail in feature extraction. RSAFormer also introduces a region self-attention enhancement module (RSA) to acquire more accurate feature information and foster a stronger interplay between low-level and high-level features. This module enhances uncertain areas to extract more precise boundary information, these areas being signified by regional context. Extensive experiments were conducted on five prevalent polyp datasets to demonstrate RSAFormer's proficiency. It achieves 92.2% and 83.5% mean Dice on Kvasir and ETIS, respectively, which outperformed most of the state-of-the-art models.

Genomic Landscape of RTK/RAS Pathway and Tumor Immune Infiltration as Prognostic Indicator of Lung Adenocarcinoma.

The RTK/RAS pathway is an oncogenic signaling pathway for which many targeted drugs have been developed; however, survival remains poor. A combination of targeted therapy and immunotherapy has emerged as an option for improving cancer treatment responses. In this study, on the basis of the expression, survival, single nucleotide variation (SNV), copy number variation (CNV), and methylation data of lung adenocarcinoma (LUAD) from The Cancer Genome Atlas database, we comprehensively analyzed the genomic changes in the RTK/RAS pathway and their associations with tumor-infiltrating lymphocytes (TIL) and prognosis in LUAD to provide the genomics landscape of RTK/RAS with TIL and prognosis. We found that two rarely mutated genes, mitogen-activated protein kinase kinase 1 and insulin-like growth factor 1 receptor, were significantly associated with the worse survival of patients with LUAD. Patients with LUAD and co-mutation of KRAS proto-oncogene (KRAS) and neurofibromin 1 genes had worse survival, and the underlying mechanism could be insufficient for protein synthesis and intracellular signal deactivation. Methylation of the Rac family small GTPase 1 (RAC1) was associated with better survival. The SNVs of the top mutated genes, including epidermal growth factor receptor (12.7%), neurotrophic receptor tyrosine kinase 3 (7.8%), erb-b2 receptor tyrosine kinase 4 (8.5%), and KRAS (29.6%), were associated with T cell exhaustion in LUAD. To construct nomograms, we further screened the genes whose genomic changes were closely associated with survival and immune infiltration. The nomograms performed well in predicting disease-specific survival (DSS) with a concordance index of 0.7 (0.589, 0.811) and overall survival with a concordance index of 0.689 (0.603, 0.775) in test set; they also showed good correspondence between actual and ideal nomogram predictions. Tumor stage, RAC1 methylation, and type 1 regulatory T cells greatly contributed to DSS and OS nomograms. In summary, we provided a comprehensive genomic profile of the RTK/RAS pathway in LUAD and its association with immune cell infiltration and prognosis of LUAD. This profile would serve as a basis for developing better therapeutic strategies, improving patient prognosis, and understanding the mechanisms of immune disturbance from the perspective of oncogenic pathways of LUAD.