Search

Nuclear energy is used more and more widely as a clean energy source, but nuclear energy facilities are risky, and when a nuclear leak occurs, it is necessary to detect equipment in a nuclear radiation environment. In the nuclear radiation environment, due to the impact of high-energy particles on the camera sensor, the collected image contains a lot of radiation noise, which greatly reduces the visual perception of the image. Aiming at the problem that radiation noise reduces image quality, a radiation image compound filtering algorithm combining median filtering and multi-frame average filtering is proposed based on radiation noise characteristics. Compared with several common filtering algorithms, the radiation noise image is filtered under the same radiation dose and achieve the highest Peak Signal Noise Rate (PSNR) and Structural Similarity (SSIM), and compared with the multi-frame average filtering method, the number of image frames required by the algorithm in this paper is greatly reduced. Experimental results show that the algorithm can effectively eliminate radiation noise and is more suitable for image filtering in radiation environment.

Background: Small cell lung cancer (SCLC) transformation had previously been reported mainly in epidermal growth factor receptor (EGFR) mutant adenocarcinoma. However, the underlying genomic profile remains unclear. Our study aimed to find the evolution and genotypic characteristic of SCLC transformation. Methods: Thirty-one SCLC transformation patients who were initially diagnosed as non-small cell lung cancer (NSCLC) patients were included. Whole exome sequencing (WES) of both primary and transformed re-biopsy lesions was conducted on 12 patients. Clinical characteristics were analyzed using R software (v.3.6.1). Results: Our study included 31 patients, of whom, three had lung squamous cell carcinoma, 6 patients did not carry EGFR mutations, and 30 patients received chemotherapy for SCLCs. The disease control rate (DCR) was 96.7%, and the median progression-free survival (PFS) was 4.03 months. The median time to transformation was 33.07 months, and the median overall survival (OS) was 62.08 months. Somatic mutation analysis showed that besides TP53, RB1, and EGFR, there was a high occurrence of mutations to CSMD3 and ADAMTS19, especially in the EGFR-wild type (EGFR-wt) group. Concerning mutational signature, the EGFR-mutant (EGFR-mut) transformed group favored an apolipoprotein B (APOBEC) mRNA editing catalytic polypeptide-like-associated mutation pattern (P = 0.16). DNA damage repair (DDR)-related signatures were significantly enriched in the EGFR-wt transformed group (P = 0.034). Additionally, clonal evolution analysis revealed that all patients had the same main trunk genes in the phylogenetic tree. Transformed SCLCs are not sensitive to immunotherapy, possibly due to increased tumor heterogeneity. Conclusions: Our results indicate that the EGFR-wt patients could also transform to SCLCs, but they have different genetic features with EGFR-mut patients. SCLC-transformed patients respond to classical chemotherapy and have a better prognosis than those with classical SCLCs.