美国白人与中国人原发性结直肠癌病理特点对照

目的　比较研究美国白人与中国患者原发性结直肠癌 (CRC)的病理学特征。方法　回顾性分析 1990～ 2 0 0 0年美国克里夫兰佛罗里达临床中心结直肠外科 6 90例和中国第一军医大学南方医院普外科 870例连续完整的CRC患者资料。结果　美国白人与中国患者比较 :36 .3%比 2 6 .0 %的肿瘤位于近侧 (右侧 )结肠 (P <0 .0 0 1) ,6 3.7%比 74 .0 %位于远侧 (左侧 )结肠 (P <0 .0 0 1) ;中位年龄 72比5 4岁 (P <0 .0 0 1) ;腺癌 84 .3%比 85 .3% ,黏液腺癌 13.6 %比 11.4 % ,印戒细胞癌 1.5 %比 2 .7% ,腺鳞癌 0 .6 %比 0 .6 % (P >0 .0 5 ) ;高分化癌 16 .7%比 4 0 .2 % ,中分化 6 1.9%比 4 8.5 % ,低分化 2 1.4 %比11.3% (P <0 .0 0 1) ,近侧结肠癌中低分化比例在两种族患者中均较高 ;TNM分期 ,0期 5 .4 %比 1.3%(P <0 .0 0 1) ,Ⅰ期 34.8%比 18.9% (P <0 .0 0 1) ,Ⅱ期 2 9.9%比 33.2 % (P >0 .0 5 ) ,Ⅲ期 2 0 .3%比2 2 .8% (P >0 .0 5 ) ,Ⅳ期 9.6 %比 17.0 % (P <0 .0 0 1)。进展期癌更常见于近侧结肠和中国患者。结论　美国白人易患近侧结肠癌 ,中国患者远侧结直肠癌更常见 ,进展期比例高 ,年龄明显小。低分化和进展期癌更常见于近侧结肠。     

Molecular cloning, chromosomal location, and tissue-specific expression of the murine cathepsin G gene

We previously have characterized a cluster of genes encoding cathepsin G (CG) and two other CG-like hematopoietic serine proteases, CGL-1 and CGL-2, on human chromosome 14. In this report, we clone and characterize a novel, related murine hematopoietic serine protease gene using human CG (hCG) cDNA as the probe. This murine gene spans approximately 2.5 kb of genomic DNA, is organized into five exons and four introns, and bears a high degree of homology to hCG at both nucleic acid (73%) and deduced amino acid (66%) levels. The predicted cDNA contains an open reading frame of 783 nucleotides that encodes a nascent protein of 261 amino acids. Processing of a putative signal (pre) peptide of 18 residues and an activation (pro) dipeptide would generate a mature enzyme of approximately 27 Kd that has an estimated pI of 12.0. Conserved residues at His44, Asp88, and Ser181 form the characteristic catalytic triad of the serine protease superfamily. The gene is tightly linked to the CTLA-1 locus on murine chromosome 14, where the serine protease genes mCCP1-4 are clustered. Expression of this gene is detected only in the bone marrow and is restricted to a small population of early myeloid cells. These findings are consistent with the identification of the gene encoding murine CG.     

Characterization of CD33 as a new member of the sialoadhesin family of cellular interaction molecules

CD33 is a member of the Ig superfamily that is restricted to cells of the myelomonocytic lineage but whose functions and binding properties are unknown. It shares sequence similarity with sialoadhesin, CD22, and the myelin-associated glycoprotein, which constitute the Sialoadhesin family of sialic acid-dependent cell adhesion molecules. In the present study, we show that CD33 is a fourth member of this family. As a model for sialic acid-dependent binding, human erythrocytes were derivatized with N-acetylneuraminic acid (NeuAc) in different linkages. A recombinant soluble form of CD33, Fc-CD33, bound red blood cells with a specificity similar to that of sialoadhesin, preferring NeuAc alpha 2,3Gal in N- and O-glycans over NeuAc alpha 2,6Gal in N-glycans. Fc- CD33 also bound selectively to the myeloid cell lines HL-60 and U937. However, CD33 was unable to mediate cell binding after transient expression in COS cells, despite high levels of surface expression. Pretreatment of the CD33-transfected cells with sialidase rendered them capable of mediating sialic acid-dependent binding. These results show that CD33 can function as a sialic acid-dependent cell adhesion molecule and that binding can be modulated by endogenous sialoglycoconjugates when CD33 is expressed in a plasma membrane.     

Effects of phonemic awareness instruction on the encoding skills of children with severe speech impairment

Purpose: To examine the effects of phoneme-grapheme correspondence and phonemic awareness instruction on the encoding abilities of three pre-reading children with severe speech impairment (SSI). Method: Using a single subject multiple baseline design across behaviours and participants, children received phoneme-grapheme awareness instruction followed by instruction in segmenting, manipulating, and encoding consonant-vowel-consonant (CVC) pseudowords. Results: Generalization occurred to encoding of novel CVC pseudo- and real words for two of the three participants. Conclusions: Results suggest that phoneme-grapheme correspondence and phonemic awareness instruction is effective in developing encoding skills in children with SSI. Findings are consistent with those for other at-risk children.     

Evaluation of federated learning variations for COVID-19 diagnosis using chest radiographs from 42 US and European hospitals

ObjectiveFederated learning (FL) allows multiple distributed data holders to collaboratively learn a shared model without data sharing. However, individual health system data are heterogeneous. “Personalized” FL variations have been developed to counter data heterogeneity, but few have been evaluated using real-world healthcare data. The purpose of this study is to investigate the performance of a single-site versus a 3-client federated model using a previously described Coronavirus Disease 19 (COVID-19) diagnostic model. Additionally, to investigate the effect of system heterogeneity, we evaluate the performance of 4 FL variations.Materials and methodsWe leverage a FL healthcare collaborative including data from 5 international healthcare systems (US and Europe) encompassing 42 hospitals. We implemented a COVID-19 computer vision diagnosis system using the Federated Averaging (FedAvg) algorithm implemented on Clara Train SDK 4.0. To study the effect of data heterogeneity, training data was pooled from 3 systems locally and federation was simulated. We compared a centralized/pooled model, versus FedAvg, and 3 personalized FL variations (FedProx, FedBN, and FedAMP).ResultsWe observed comparable model performance with respect to internal validation (local model: AUROC 0.94 vs FedAvg: 0.95, P = .5) and improved model generalizability with the FedAvg model (P < .05). When investigating the effects of model heterogeneity, we observed poor performance with FedAvg on internal validation as compared to personalized FL algorithms. FedAvg did have improved generalizability compared to personalized FL algorithms. On average, FedBN had the best rank performance on internal and external validation.ConclusionFedAvg can significantly improve the generalization of the model compared to other personalization FL algorithms; however, at the cost of poor internal validity. Personalized FL may offer an opportunity to develop both internal and externally validated algorithms.