The establishment of image correspondence through robust image registration is critical to many clinical tasks such as image fusion, organ atlas creation, and tumor growth monitoring, and is a very challenging problem. Since the beginning of the recent deep learning renaissance, the medical imaging research community has developed deep learning based approaches and achieved the stateof-the-art in many applications, including image registration. The rapid adoption of deep learning for image registration applications over the past few years necessitates a comprehensive summary and outlook, which is the main scope of this survey. This requires placing a focus on the different research areas as well as highlighting challenges that practitioners face. This survey, therefore, outlines the evolution of deep learning based medical image registration in the context of both research challenges and relevant innovations in the past few years. Further, this survey highlights future research directions to show how this field may be possibly moved forward to the next level.

Low-dose computed tomography (LDCT) has attracted major attention in the medical imaging field, since CT-associated X-ray radiation carries health risks for patients. The reduction of the CT radiation dose, however, compromises the signal-to-noise ratio, which affects image quality and diagnostic performance. Recently, deep-learning-based algorithms have achieved promising results in LDCT denoising, especially convolutional neural network (CNN) and generative adversarial network (GAN) architectures. This paper introduces a conveying path-based convolutional encoder-decoder (CPCE) network in 2-D and 3-D configurations within the GAN framework for LDCT denoising. A novel feature of this approach is that an initial 3-D CPCE denoising model can be directly obtained by extending a trained 2-D CNN, which is then fine-tuned to incorporate 3-D spatial information from adjacent slices. Based on the transfer learning from 2-D to 3-D, the 3-D network converges faster and achieves a better denoising performance when compared with a training from scratch. By comparing the CPCE network with recently published work based on the simulated Mayo data set and the real MGH data set, we demonstrate that the 3-D CPCE denoising model has a better performance in that it suppresses image noise and preserves subtle structures.

No abstract

Pancreatic ductal adenocarcinoma (PDAC) still ranking 4th in the order of fatal tumor diseases is characterized by a profound tumor stroma with high numbers of tumor-associated macrophages (TAMs). Driven by environmental factors, monocytes differentiate into M1-or M2-macrophages, the latter commonly regarded as being protumorigenic. Because a detailed analysis of TAMs in human PDAC development is still lacking, freshly isolated PDAC-derived TAMs were analyzed for their phenotype and impact on epithelial-mesenchymal-transition (EMT) of benign (H6c7) and malignant (Colo357) pancreatic ductal epithelial cells. TAMs exhibited characteristics of M1-macrophages (expression of HLA-DR, IL-1b, or TNF-a) and M2-macrophages (expression of CD163 and IL-10). In the presence of TAMs, H6c7, and Colo357 cells showed an elongated cell shape along with an increased expression of mesenchymal markers such as vimentin and reduced expression of epithelial E-cadherin. Similar to TAMs, in vitro generated M1-and M2-macrophages both mediated EMT in H6c7 and Colo357 cells. M1-macrophages acquired M2-characteristics during coculture that could be prevented by GM-CSF treatment. However, M1-macrophages still potently induced EMT in H6c7 and Colo357 cells although lacking M2-characteristics. Overall, these data demonstrate that TAMs exhibit anti-as well as proinflammatory properties that equally contribute to EMT induction in PDAC initiation and development.Pancreatic ductal adenocarcinoma (PDAC) is a highly malignant tumor, still with a dismal prognosis. In Western countries, PDAC ranks 4th in the order of death related tumor diseases with a still increasing prevalence.1 Commonly, PDAC is detected in an advanced stage when the tumor has already metastasized so that therapeutic options are very limited, in line with low overall 5-year survival rates of <5%. 2One hallmark of PDAC which is supposed to originate from the ductal epithelium is the pronounced tumor stroma. This marked stromal enrichment is already present in