Deep learning-derived spatial organization features on histology images predicts prognosis in colorectal liver metastasis patients after hepatectomy.

Histopathological images of colorectal liver metastases (CRLM) contain rich morphometric information that may predict patients' outcomes. However, to our knowledge, no study has reported any practical deep learning framework based on the histology images of CRLM, and their direct association with prognosis remains largely unknown. In this study, we developed a deep learning-based framework for fully automated tissue classification and quantification of clinically relevant spatial organization features (SOFs) in H&E-stained images of CRLM. The SOFs based risk-scoring system demonstrated a strong and robust prognostic value that is independent of the current clinical risk score (CRS) system in independent clinical cohorts. Our framework enables fully automated tissue classification of H&E images of CRLM, which could significantly reduce assessment subjectivity and the workload of pathologists. The risk-scoring system provides a time- and cost-efficient tool to assist clinical decision-making for patients with CRLM, which could potentially be implemented in clinical practice.

A computational study on the influence of aortic valve disease on hemodynamics in dilated aorta.

A computational hemodynamics method was employed to investigate how the morphotype and functional state of aortic valve would affect the characteristics of blood flow in aortas with pathological dilation, especially the intensity and distribution of flow turbulence. Two patient-specific aortas diagnosed to have pathological dilation of the ascending segment while differential aortic valve conditions (i.e., one with a stenotic and regurgitant RL bicuspid aortic valve (RL-BAV), whereas the other with a quasi-normal tricuspid aortic valve (TAV)) were studied. When building the computational models, in addition to in vivo data-based reconstruction of geometrical model and boundary condition setting, the large eddy simulation method was adopted to quantify potential flow turbulence in the aortas. Obtained results revealed the presence of complex flow patterns (denoted by time-varying changes in vortex structure), flow turbulence (indicated by high turbulent eddy viscosity (TEV)), and regional high wall shear stress (WSS) in the ascending segment of both aortas. Such hemodynamic characteristics were significantly augmented in the aorta with RL-BAV. For instance, the space-averaged TEV in late systole and the wall area exposed to high time-averaged WSS (judged by WSS> two times of the mean WSS in the entire aorta) in the ascending aortic segment were increased by 176% and 465%, respectively. Relatively, flow patterns in the descending aortic segment were less influenced by the aortic valve disease. These results indicate that aortic valve disease has profound impacts on flow characteristics in the ascending aorta, especially the distribution and degree of high WSS and flow turbulence.

Differential microRNA expression by Solexa sequencing in the sera of ovarian cancer patients.

MicroRNAs are a class of small noncoding RNA which play important regulatory roles in a variety of cancers. MiRNA-specific expression profiles have been reported for several pathological conditions. In this study, we combined large scale parallel Solexa sequencing to identify 11 up-regulated miRNAs and 19 down-regulated miRNAs with computational techniques in the sera of ovarian cancer patients while using healthy serum as the control. Among the above, four miRNAs (miR-22, miR-93, miR-106b, miR-451) were validated by quantitative RT-PCR and found to be significantly aberrantly expressed in the serum of ovarian cancer patients (P<0.05). There were no significant differences between samples from cancer stage I/II and III/IV. However, the levels of miR-106b (p=0.003) and miR-451 (p=0.007) were significantly different in those patients under and over 51 yearsof age. MiR-451 and miR-93 were also specific when analyzed with reference to different levels of CA125. This study shows that Solexa sequencing provides a promising method for cancer-related miRNA profiling, and selectively expressed miRNAs could be used as potential serum-based biomarkers for ovarian cancer diagnosis.

Modeling of Fusarium redolens Dzf2 mycelial growth kinetics and optimal fed-batch fermentation for beauvericin production.

Beauvericin (BEA) is a cyclic hexadepsipeptide mycotoxin with notable phytotoxic and insecticidal activities. Fusarium redolens Dzf2 is a highly BEA-producing fungus isolated from a medicinal plant. The aim of the current study was to develop a simple and valid kinetic model for F. redolens Dzf2 mycelial growth and the optimal fed-batch operation for efficient BEA production. A modified Monod model with substrate (glucose) and product (BEA) inhibition was constructed based on the culture characteristics of F. redolens Dzf2 mycelia in a liquid medium. Model parameters were derived by simulation of the experimental data from batch culture. The model fitted closely with the experimental data over 20-50 g l(-1) glucose concentration range in batch fermentation. The kinetic model together with the stoichiometric relationships for biomass, substrate and product was applied to predict the optimal feeding scheme for fed-batch fermentation, leading to 54% higher BEA yield (299 mg l(-1)) than in the batch culture (194 mg l(-1)). The modified Monod model incorporating substrate and product inhibition was proven adequate for describing the growth kinetics of F. redolens Dzf2 mycelial culture at suitable but not excessive initial glucose levels in batch and fed-batch cultures.