A Semi-Supervised Learning Framework for Classifying Colorectal Neoplasia Based on the NICE Classification.

Labelling medical images is an arduous and costly task that necessitates clinical expertise and large numbers of qualified images. Insufficient samples can lead to underfitting during training and poor performance of supervised learning models. In this study, we aim to develop a SimCLR-based semi-supervised learning framework to classify colorectal neoplasia based on the NICE classification. First, the proposed framework was trained under self-supervised learning using a large unlabelled dataset; subsequently, it was fine-tuned on a limited labelled dataset based on the NICE classification. The model was evaluated on an independent dataset and compared with models based on supervised transfer learning and endoscopists using accuracy, Matthew's correlation coefficient (MCC), and Cohen's kappa. Finally, Grad-CAM and t-SNE were applied to visualize the models' interpretations. A ResNet-backboned SimCLR model (accuracy of 0.908, MCC of 0.862, and Cohen's kappa of 0.896) outperformed supervised transfer learning-based models (means: 0.803, 0.698, and 0.742) and junior endoscopists (0.816, 0.724, and 0.863), while performing only slightly worse than senior endoscopists (0.916, 0.875, and 0.944). Moreover, t-SNE showed a better clustering of ternary samples through self-supervised learning in SimCLR than through supervised transfer learning. Compared with traditional supervised learning, semi-supervised learning enables deep learning models to achieve improved performance with limited labelled endoscopic images.

Low LINC01272 predicts poor prognosis of non-small cell lung cancer and its biological function in tumor cells by inhibiting miR-1303.

Non-small cell lung cancer (NSCLC) is a malignant tumor associated with poor prognosis. The clinical value of long non-coding RNAs (lncRNAs) in the pathomechanism of various types of human malignancy has attracted increasing attention. The present study aimed to investigate the expression of LINC01272 in NSCLC and to determine its prognostic value and biological role. Tumor and adjacent non-tumor tissues from 108 patients with NSCLC and NSCLC cell lines were used in this study. The expression levels of LINC01272 and microRNA (miR)-1303 in tissues of patients and NSCLC cell lines were evaluated by reverse transcription quantitative PCR. The relationship between LINC01272 and the overall survival of patients with NSCLC was analyzed by Kaplan-Meier survival curve and log-rank test. Cox regression analysis confirmed the prognostic value of LINC01272 in patients with NSCLC. Cell Counting Kit-8 assay was used to evaluate the proliferation of NSCLC cells. The migration and invasion of NSCLC cells were determined using Transwell assays. The interaction between LINC01272 and miR-1303 in NSCLC was confirmed by dual-luciferase reporter assay. LINC01272 downregulation in NSCLC tissues was associated with worse overall survival in patients based on bioinformatics analysis. Furthermore, LINC01272 expression, which was decreased in NSCLC tumor tissues and NSCLC cells, was considered as an independent prognostic biomarker in NSCLC. In addition, LINC01272 overexpression inhibited NSCLC cell proliferation, migration and invasion. miR-1303 expression, which was increased in tumor tissues, was sponged by LINC01272 and negatively correlated with LINC01272 expression. miR-1303 expression reversed the inhibitory effects of LINC01272 on NSCLC cell function. In summary, the findings from this study suggested that LINC01272 expression, which was decreased in NSCLC tumor tissues and NSCLC cells, may be used as an independent prognostic biomarker for patients with NSCLC and that its overexpression may suppress NSCLC cell proliferation, migration and invasion by inhibiting miR-1303.

Lung inflammation caused by long-term exposure to titanium dioxide in mice involving in NF-κB signaling pathway.

Titanium dioxide nanoparticles (TiO2 NPs) are used in many fields, such as paints, medicine additives, food additives, sunscreens, and agriculture. The aim of this study was to investigate the mechanism behind the formation of inflammation induced by TiO2 NPs. ICR mice were exposed to TiO2 NPs through intragastric administration at 2.5, 5, and 10 mg/kg body weight every day for 90 consecutive days. The experiment suggested that long-term exposure to TiO2 NPs resulted in an obvious inflammatory response in mice lung tissues, which led to a thickened alveoli septum, lung hyperemia, and titanium accumulation. Furthermore, our results show that TiO2 NPs exposure remarkably altered the expression of inflammation-related cytokines, with increases in proinflammatory cytokines-such as nucleic factor-κB, interferon-α, interferon-ß, interleukin-1ß, interleukin-6, cyclo-oxygen-ase, interleukin-8, interferon-inducible protein-10, and platelet-derived growth factor AB-and decreases in anti-inflammatory cytokines-such as inhibitor of NF-κB suppressor of cytokine signaling 1, endothelin 1, peroxisome proliferators-activated receptors-γ, and peroxisome proliferators-activated receptors coactivator-1α. This finding indicated that TiO2 NPs cause lung inflammation in mice after intragastric administration, primarily through the NF-κB signaling pathways. Therefore, more attention should be placed on the application of TiO2 NPs and their potential long-term effects, especially in human beings. © 2016 Wiley Periodicals, Inc. J Biomed Mater Res Part A: 105A: 720-727, 2017.