Search

With the renaissance of deep learning, automatic diagnostic algorithms for computed tomography (CT) have achieved many successful applications. However, they heavily rely on lesion-level annotations, which are often scarce due to the high cost of collecting pathological labels. On the other hand, the annotated CT data, especially the 3-D spatial information, may be underutilized by approaches that model a 3-D lesion with its 2-D slices, although such approaches have been proven effective and computationally efficient. This study presents a multiview contrastive network (MVCNet), which enhances the representations of 2-D views contrastively against other views of different spatial orientations. Specifically, MVCNet views each 3-D lesion from different orientations to collect multiple 2-D views; it learns to minimize a contrastive loss so that the 2-D views of the same 3-D lesion are aggregated, whereas those of different lesions are separated. To alleviate the issue of false negative examples, the uninformative negative samples are filtered out, which results in more discriminative features for downstream tasks. By linear evaluation, MVCNet achieves state-of-the-art accuracies on the lung image database consortium and image database resource initiative (LIDC-IDRI) (88.62%), lung nodule database (LNDb) (76.69%), and TianChi (84.33%) datasets for unsupervised representation learning. When fine-tuned on 10% of the labeled data, the accuracies are comparable to the supervised learning models (89.46% versus 85.03%, 73.85% versus 73.44%, 83.56% versus 83.34% on the three datasets, respectively), indicating the superiority of MVCNet in learning representations with limited annotations. Our findings suggest that contrasting multiple 2-D views is an effective approach to capturing the original 3-D information, which notably improves the utilization of the scarce and valuable annotated CT data.

Introduction: The tumor immune environment and immune-related genes are instrumental in the development, progression, and prognosis of bladder cancer (BLCA). This study sought to pinpoint key immune-related genes influencing BLCA prognosis and decipher their mechanisms of action., Methods and Results: We analyzed differentially expressed genes (DEGs) between high- and low- tumor mutational burden (TMB) groups. Subsequently, we constructed a reliable prognostic model based on immune-related gene pairs (IRGPs) and analyzed DEGs between high- and low-risk groups. A total of 22 shared DEGs were identified across differential TMB and IRGPs-derived risk groups in BLCA patients. Through univariate Cox and multivariate Cox analyses, we highlighted five genes - FLRT2, NTRK2, CYTL1, ZNF683, PRSS41 - significantly correlated with BLCA patient prognosis. Notably, the FLRT2 gene emerged as an independent prognostic factor for BLCA, impacting patient prognosis via modulation of macrophage infiltration in immune microenvironment. Further investigation spotlighted methylation sites - cg25120290, cg02305242, and cg01832662 - as key regulators of FLRT2 expression., Discussion: These findings identified pivotal prognostic genes in BLCA and illuminated the intricate mechanisms dictating patient prognosis. This study not only presents a novel prognostic marker but also carves out potential avenues for immunotherapy and targeted therapeutic strategies in BLCA. By demystifying the profound impact of immune-related genes and the tumor immune environment, this study augments the comprehension and prognostic management of bladder cancer., Competing Interests: Authors JZ and QC were employed by the company Thorgene Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Tao, Yu, Cong, Li, Zhu, Ding, Chen and Cai.)

Understanding the key factors in the tumor microenvironment (TME) that affect the prognosis of gliomas is crucial. In this study, we sought to uncover the prognostic significance of immune cells and immune-related genes in the TME of gliomas. We incorporated data of 970 glioma patient samples from the Chinese Glioma Genome Atlas (CGGA) database as the training set, and an additional set of 666 samples from The Cancer Genome Atlas (TCGA) database served as the validation set. From our analysis, we identified 21 immune-related differentially expressed genes (DEGs) in the TME, which holds implications for glioma prognosis. Based on these genes, we constructed a prognostic risk model on the 21 genes. The prognostic risk model demonstrated robust performance with an area under the curve (AUC) value of 0.848. Notably, the risk score derived from the model emerged as an independent prognostic factor of gliomas, with high risk scores indicative of an unfavorable prognosis. Furthermore, we observed that high infiltration levels of certain immune cells, namely, activated dendritic cells, M0 macrophages, M2 macrophages, and regulatory T cells (Tregs), correlated with an unfavorable glioma prognosis. In conclusion, our findings suggested that the TME of gliomas harbored a distinct immune-associated signature, comprising 21 immune-related genes and specific immune cells. These elements significantly influence the prognosis and present potential as novel indicators in the clinical assessment of glioma patient outcomes., Competing Interests: Authors JZ and QC were employed by ThorGene Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Tao, Zhu, Yu, Cong, Li, Cai and Chen.)

Liver Cancer is a threat to human health and life over the world. The key to reduce liver cancer incidence is to identify high-risk populations and carry out individualized interventions before cancer occurrence. Building predictive models based on machine learning algorithms is an effective and economical way to forecast potential liver cancers. However, since the dataset is usually extremely skewed (negative samples are much more than positive samples), machine learning models suffer from severe bias and make unreliable predictions. In this paper, we systematically evaluate existing approaches in tackling class-imbalance problem and introduce two undersampling methods. The first is based on K-means++, where robust clustering centers are appointed as negative samples. The second is based on learning vector quantization, which considers diagnostic labels during clustering, and the prototypes are used as negative data. In this way, positive and negative samples are rebalanced. The algorithm is applied to five-year liver cancer prediction in Early Diagnosis and Treatment of Urban Cancer project in China. We achieve an AUC of 0.76 when no clinical measure except for epidemiological information is used. Experimental results show the advantage of our method over existing oversampling, undersampling, ensemble algorithms, and state-of-the-art outlier detection algorithms. This work explores a feasible and practical roadmap to tackle skewed medical data in cancer prediction and benefits applications targeted to human health and well-being., (Copyright © 2022 Elsevier B.V. All rights reserved.)