RESUMO
BACKGROUND: Bone age assessment assists physicians in evaluating the growth and development of children. However, deep learning methods for bone age estimation do not currently incorporate differential features obtained through comparisons with other bone atlases. OBJECTIVE: To propose a more accurate method, Delta-Age-Sex-AdaIn (DASA-net), for bone age assessment, this paper combines age and sex distribution through adaptive instance normalization (AdaIN) and style transfer, simulating the process of visually comparing hand images with a standard bone atlas to determine bone age. MATERIALS AND METHODS: The proposed Delta-Age-Sex-AdaIn (DASA-net) consists of four modules: BoneEncoder, Binary code distribution, Delta-Age-Sex-AdaIn, and AgeDecoder. It is compared with state-of-the-art methods on both a public Radiological Society of North America (RSNA) pediatric bone age prediction dataset (14,236 hand radiographs, ranging from 1 to 228 months) and a private bone age prediction dataset from Zigong Fourth People's Hospital (474 hand radiographs, ranging from 12 to 218 months, 268 male). Ablation experiments were designed to demonstrate the necessity of incorporating age distribution and sex distribution. RESULTS: The DASA-net model achieved a lower mean absolute deviation (MAD) of 3.52 months on the RSNA dataset, outperforming other methods such as BoneXpert, Deeplasia, BoNet, and other deep learning based methods. On the private dataset, the DASA-net model obtained a MAD of 3.82 months, which is also superior to other methods. CONCLUSION: The proposed DASA-net model aided the model's learning of the distinctive characteristics of hand bones of various ages and both sexes by integrating age and sex distribution into style transfer.
RESUMO
A new approach is proposed that identifies three different zones of the Si-rich network structure (the cellular structure) in laser powder bed fused (LPBF) AlSi10Mg alloy, based on the variation in morphology, grain growth transition, and melt pool solidification conditions. The three identified zones are denoted in the present work as the liquid solidification zone (LSZ), the mushy solidification zone (MSZ), and the heat affected zone (HAZ). The LSZ is the result of liquid-solid transformation, showing small planar growth at the boundary and large cellular growth in the center, while the MSZ is related to a semisolid reaction, and the HAZ arises from a short-time aging process. The boundary between the LSZ and MSZ is identified by the change of grain growth direction and the Si-rich network advancing direction. The boundary between MSZ and HAZ is identified by the start of the breakdown of the Si-rich network. In addition, it is found that the fracture is generated in and propagates along the HAZ during tensile tests.
RESUMO
Background: Immunogenic cell death (ICD) is related to cancer prognosis, which has a synergic effect in combination with chemotherapy or immunotherapy. Yet, the relationship between ICD and osteosarcoma remained unclear. Materials and methods: Three osteosarcoma datasets including therapeutically applicable research to generate effective treatments (TARGET), GSE126209 and GSE21257 datasets were included. A protein-protein interaction network was constructed based on ICD-related genes. We performed unsupervised consensus clustering to classify molecular subtypes (clusters). Survival analysis, Estimation of stromal and immune cells in malignant tumour tissues using expression data (ESTIMATE), Cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT), and differential analysis were employed to characterize the molecular differences between different clusters. Univariate Cox regression analysis was conducted to confirm prognostic genes. Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was used to demonstrate the aberrant expression of ICD-correlated signature genes in osteosarcoma. A series of cellular experiments, including cell counting kit-8 (CCK-8), transwell, and flow cytometry, were used to demonstrate the regulatory role of key genes in the ICD model on the malignant phenotype of osteosarcoma. Results: Three clusters (cluster1, 2, 3) were constructed and they showed distinct overall survival and immune infiltration. ICD-related genes were highly expressed in cluster1. Moreover, Cluster1 had the best prognosis, high immune score and high expression of human leukocyte antigen (HLA)-related genes. TLR4, LY96, IFNGR1, CD4, and CASP1 were identified as prognostic genes for establishing an ICD-related risk signature. According to the risk signature, two risk groups (high and low risks) showing differential prognosis and response to immunotherapy. The low risks group had a better prognosis but was not sensitive to immunotherapy. Molecular assays verified that prognostic genes were abnormally under-expressed in osteosarcoma. Cellular assays demonstrated that LY96, the most significantly down-regulated gene in osteosarcoma, inhibited the migration, invasion, and proliferation phenotypes of osteosarcoma cells and prolonged the cell cycle. Analysis of oxidative stress related pathway enrichment in tumor microenvironment was conducted by single-sample gene set enrichment analysis (ssGSEA). Conclusions: This study demonstrated the prognostic significance of ICD-correlated genes in osteosarcoma patients. The five-gene risk signature facilitate prognostic evaluation and prediction of osteosarcoma patients' response to immunotherapy. The risk signature also offered a possibility for the exploit of novel ICD-related treatment.