No abstract

Lung cancer is the leading cause of cancer-related death worldwide. Early diagnosis of pulmonary nodules in Computed Tomography (CT) chest scans provides an opportunity for designing effective treatment and making financial and care plans. In this paper, we consider the problem of diagnostic classification between benign and malignant lung nodules in CT images, which aims to learn a direct mapping from 3D images to class labels. To achieve this goal, four two-pathway Convolutional Neural Networks (CNN) are proposed, including a basic 3D CNN, a novel multi-output network, a 3D DenseNet, and an augmented 3D DenseNet with multi-outputs. These four networks are evaluated on the public LIDC-IDRI dataset and outperform most existing methods. In particular, the 3D multi-output DenseNet (MoD-enseNet) achieves the state-of-the-art classification accuracy on the task of end-to-end lung nodule diagnosis. In addition, the networks pretrained on the LIDC-IDRI dataset can be further extended to handle smaller datasets using transfer learning. This is demonstrated on our dataset with encouraging prediction accuracy in lung nodule classification.

No abstract

The delivery of high dose radiotherapy to tumors is often limited by the proximity of the surrounding radiosensitive normal tissues, even using modern techniques such as intensity modulated radiation therapy (IMRT). Previous studies have reported that placement of a spacer can effectively displace normal tissues. So that they are some distance away from the lesion, thus allowing for the safe delivery of high-dose radiation. The application of radioprotective spacers was first reported 30 years ago regarding radiotherapy of tongue and abdominal cancers; more recently, they are increasingly being used in prostate cancer. This review focuses on the published data concerning the features of different types of spacers and their application in various tumor sites. Placement-related complications and the cost-effectiveness of the spacers are also discussed. With the increasing use of high-precision radiotherapy in clinical practice, especially the paradigm-changing stereotactic body radiotherapy (SBRT), more robust studies are warranted to further establish the role of radioprotective spacers through materials development and novel placement techniques.

Mutations in ATP8B1 or ATP11C (members of P4-type ATPases) cause progressive familial intrahepatic cholestasis type 1 in human or intrahepatic cholestasis in mice. Transmembrane protein 30A (TMEM30A), a β-subunit, is essential for the function of ATP8B1 and ATP11C. However, its role in the etiology of cholestasis remains poorly understood. To investigate the function of TMEM30A in bile salt (BS) homeostasis, we developed Tmem30a liver-specific knockout (LKO) mice. Tmem30a LKO mice experienced hyperbilirubinemia, hypercholanemia, inflammatory infiltration, ductular proliferation, and liver fibrosis. The expression and membrane localization of ATP8B1 and ATP11C were significantly reduced in Tmem30a LKO mice, which correlated with the impaired expression and localization of BS transporters, such as OATP1A4, OATP1B2, NTCP, BSEP, and MRP2. The proteasome inhibitor bortezomib partially restored total protein levels of BS transporters but not the localization of BS transporters in the membrane. Furthermore, the expression of nuclear receptors, including FXRα, RXRα, HNF4α, LRH-1, and SHP, was also down-regulated. A cholic acid-supplemented diet exacerbated the liver damage in Tmem30a LKO mice. TMEM30A deficiency led to intrahepatic cholestasis in mice by impairing the expression and localization of BS transporters and the expression of related nuclear receptors. Therefore, TMEM30A may be a novel genetic determinant of intrahepatic cholestasis.