Fully automated grading system for the evaluation of punctate epithelial erosions using deep neural networks.

PURPOSE: The goal was to develop a fully automated grading system for the evaluation of punctate epithelial erosions (PEEs) using deep neural networks. METHODS: A fully automated system was developed to detect corneal position and grade staining severity given a corneal fluorescein staining image. The fully automated pipeline consists of the following three steps: a corneal segmentation model extracts corneal area; five image patches are cropped from the staining image based on the five subregions of extracted cornea; a staining grading model predicts a score for each image patch from 0 to 3, and automated grading score for the whole cornea is obtained from 0 to 15. Finally, the clinical grading scores annotated by three ophthalmologists were compared with automated grading scores. RESULTS: For corneal segmentation, the segmentation model achieved an intersection over union of 0.937. For punctate staining grading, the grading model achieved a classification accuracy of 76.5% and an area under the receiver operating characteristic curve of 0.940 (95% CI 0.932 to 0.949). For the fully automated pipeline, Pearson's correlation coefficient between the clinical and automated grading scores was 0.908 (p<0.01). Bland-Altman analysis revealed 95% limits of agreement between the clinical and automated grading scores of between -4.125 and 3.720 (concordance correlation coefficient=0.904). The average time required for processing a single stained image during pipeline was 0.58 s. CONCLUSION: A fully automated grading system was developed to evaluate PEEs. The grading results may serve as a reference for ophthalmologists in clinical trials and residency training procedures.

A Fully Automated Segmentation and Morphometric Parameter Estimation System for Assessing Corneal Endothelial Cell Images.

PURPOSE: To develop a fully automated segmentation and morphometric parameter estimation system for assessing corneal endothelial cells from in vivo confocal microscopy images. DESIGN: Artificial intelligence (neural network) study. METHODS: First, a fully automated deep learning system for assessing corneal endothelial cells was developed using the development set (from 99 subjects). Second, 184 images (from 97 subjects) were used to construct the testing set to evaluate the clinical validity and usefulness of the automated segmentation and morphometric system. Third, the automatically calculated endothelial cell density (ECD) values, Topcon's cell density, and manually calculated ECD were compared. RESULTS: After slit lamp examination, 88 healthy subjects, 2 Fuchs endothelial dystrophy patients, and 7 corneal endotheliitis patients were identified among the 97 subjects in the testing set. The automatedly estimated morphometric parameters for the testing set were an average number of 234 cells, an ECD of 2592 cells/mm2, a coefficient of variation in the cell area of 32.14%, and a percentage of hexagonal cells of 54.16%. Pearson's correlation coefficient between the automated ECD and Topcon's cell density and between the manually calculated ECD and Topcon's cell density was 0.932 (P < .01) and 0.818 (P < .01), respectively. The Bland-Altman plot of Topcon's cell density and the automated ECD yielded 95% limits of agreement between 271.94 and -572.46 (concordance correlation coefficient = 0.9). CONCLUSIONS: A fully automated method for segmenting corneal endothelial cells and estimating morphometric parameters using in vivo confocal microscopy images is more efficient and accurate for assessing the normal corneal endothelium.

SFRP4 is a prognostic marker and correlated with Treg cell infiltration in pancreatic ductal adenocarcinoma.

Secreted Frizzled-Related Protein 4 (SFRP4), a member of secreted frizzled-related protein family, has been found as a vital modulator in cell proliferation, cell self-renew and apoptosis through Wnt signaling transduction pathway. In the present study, we re-analyzed the expression pattern of SFRPs in Gene Expression Omnibus (GEO) datasets and evaluated the expression of SFRP4 at protein level in both KrasG12D/+; Trp53R172H/+; Pdx1-Cre; (KPC) mice and human pancreatic ductal adenocarcinoma (PDAC) tissue. We found that the expression of SFRP4 increased gradually in PanINs and PDAC lesions in KPC mice and high expression of SFRP4 was much more common in tumor lesions compared to the adjacent non-tumor tissues. Then we performed Kaplan-Meier survival and Cox regression analysis and found that high expression of SFRP4 in the serum and tumor lesions predicted poor prognosis for pancreatic cancer patients. Furthermore, we demonstrated that SFRP4 positively correlated with FOXP3+ Treg cells infiltration while the down-regulation of SFRP4 in tumor cells impaired the production of cytokines and the recruitments of T cells. This study suggested that SFRP4 can be a novel prognostic biomarker and potential therapeutic target for pancreatic cancer.

Interleukin-8, a promising predictor for prognosis of pancreatic cancer.

AIM: To investigate the value of interleukin-8 (IL-8), a pro-inflammatory chemokine, in predicting the prognosis of pancreatic cancer. METHODS: Expression of IL-8 and its receptor CXCR1 was assessed by immunohistochemistry in pancreatic cancer and chronic pancreatitis samples. Enzyme-linked immunosorbent assay was used to detect the serum IL-8 levels in pancreatic cancer patients. Human pancreatic cancer tissues were heterotopically transplanted to the immune-deficiency mice to evaluate the effect of serum IL-8 on the tumorigenesis of the cancer samples. RESULTS: IL-8 and CXCR1 proteins were both over-expressed in pancreatic adenocarcinoma samples (55.6% and 65.4%, respectively) compared with the matched para-cancer tissues (25.9% and 12.3%, P < 0.01), or chronic pancreatitis (0% and 25%, P < 0.05). Serum IL-8 levels in pancreatic cancer patients (271.1 ± 187.7 ng/mL) were higher than in other digestive system tumors, such as gastric cancer (41.77 ± 9.11 ng/mL, P = 0.025), colorectal carcinoma (78.72 ± 80.60 ng/mL, P = 0.032) and hepatocellular carcinoma (59.60 ± 19.80 ng/mL, P = 0.016). In vivo tumorigenesis analysis further proved that tumor tissues from patients with higher serum IL-8 levels grew faster than those with lower IL-8 levels. CONCLUSION: IL-8 can be a fine serum marker for predicting the prognosis pancreatic cancer.

Recombinant adenovirus carrying the hepatocyte nuclear factor-1alpha gene inhibits hepatocellular carcinoma xenograft growth in mice.

UNLABELLED: Hepatocyte nuclear factor-1alpha (HNF1α) is one of the key transcription factors of the HNF family, which plays a critical role in hepatocyte differentiation. Substantial evidence has suggested that down-regulation of HNF1α may contribute to the development of hepatocellular carcinoma (HCC). Herein, human cancer cells and tumor-associated fibroblasts (TAFs) were isolated from human HCC tissues, respectively. A recombinant adenovirus carrying the HNF1α gene (AdHNF1α) was constructed to determine its effect on HCC in vitro and in vivo. Our results demonstrated that HCC cells and HCC tissues revealed reduced expression of HNF1α. Forced reexpression of HNF1α significantly suppressed the proliferation of HCC cells and TAFs and inhibited the clonogenic growth of hepatoma cells in vitro. In parallel, HNF1α overexpression reestablished the expression of certain liver-specific genes and microRNA 192 and 194 levels, with a resultant increase in p21 levels and induction of G(2)/M arrest. Additionally, AdHNF1α inhibited the expression of cluster of differentiation 133 and epithelial cell adhesion molecule and the signal pathways of the mammalian target of rapamycin and transforming growth factor beta/Smads. Furthermore, HNF1α abolished the tumorigenicity of hepatoma cells in vivo. Most interestingly, intratumoral injection of AdHNF1α significantly inhibited the growth of subcutaneous HCC xenografts in nude mice. Systemic delivery of AdHNF1α could eradicate the orthotopic liver HCC nodules in nonobese diabetic/severe combined immunodeficiency mice. CONCLUSION: These results suggest that the potent inhibitive effect of HNF1α on HCC is attained by inducing the differentiation of hepatoma cells into mature hepatocytes and G(2)/M arrest. HNF1α might represent a novel, promising therapeutic agent for human HCC treatment. Our findings also encourage the evaluation of differentiation therapy for tumors of organs other than liver using their corresponding differentiation-determining transcription factor.

Downregulation of Dickkopf-1 is responsible for high proliferation of breast cancer cells via losing control of Wnt/beta-catenin signaling.

AIM: To investigate the role of DKK-1/Wnt/beta-catenin signaling in high proliferation of LM-MCF-7 breast cancer cells, a sub-clone of MCF-7 cell line. METHODS: Two cell lines (MCF-7 and LM-MCF-7) with different proliferation abilities were used. LM-MCF-7 cells were transiently transfected with the pcDNA3-DKK-1 plasmid encoding the DKK-1 gene (or MCF-7 cells were transfected siRNA targeting DKK-1 mRNA). Flow cytometry analysis and 5-bromo-2'-deoxyuridine (BrdU) incorporation assay were applied to detect the cell proliferation. The expression levels of beta-catenin, phosphorylated beta-catenin, c-Myc, cyclin D1 and Survivin were examined by Western blot analysis. The regulation of Survivin was investigated by Luciferase reporter gene assay. RESULTS: Western blot and RT-PCR analysis showed that the expression level of DKK-1 was downregulated in LM-MCF-7 relative to MCF-7 cells. Flow cytometry and BrdU incorporation assay showed DKK-1 could suppress growth of breast cancer cells. Overexpression of DKK-1 was able to accelerate phosphorylation-dependent degradation of beta-catenin and downregulate the expression of beta-catenin, c-Myc, cyclin D1 and Survivin. Luciferase reporter gene assay demonstrated that Survivin could be regulated by beta-catenin/TCF4 pathway. CONCLUSION: We conclude that the downregulation of DKK-1 is responsible for the high proliferation ability of LM-MCF-7 breast cancer cells via losing control of Wnt/beta-catenin signaling pathway, in which c-Myc, cyclinD1 and Survivin serve as essential downstream effectors. Our finding provides a new insight into the mechanism of breast cancer cell proliferation.