Stereotactic body radiotherapy (SBRT) has become a standard treatment option for early stage, node negative non-small cell lung cancer (NSCLC) in patients who are either medically inoperable or refuse surgical resection. SBRT has high local control rates and a favorable toxicity profile relative to other surgical and non-surgical approaches. Given the excellent tumor control rates and increasing utilization of SBRT, recent efforts have focused on limiting toxicity while expanding treatment to increasingly complex patients. We review toxicities from SBRT for lung cancer, including central airway, esophageal, vascular (e.g., aorta), lung parenchyma (e.g., radiation pneumonitis), and chest wall toxicities, as well as radiation-induced neuropathies (e.g., brachial plexus, vagus nerve and recurrent laryngeal nerve). We summarize patient-related, tumor-related, dosimetric characteristics of these toxicities, review published dose constraints, and propose strategies to reduce such complications.

 Abstract-Convolutional neural networks (CNNs) demonstrate excellent performance in various computer vision applications. In recent years, FPGA-based CNN accelerators have been proposed for optimizing performance and power efficiency. Most accelerators are designed for object detection and recognition algorithms that are performed on low-resolution (LR) images. However, real-time image super-resolution (SR) cannot be implemented on a typical accelerator because of the long execution cycles required to generate high-resolution (HR) images, such as those used in ultra-high-definition (UHD) systems. In this paper, we propose a novel CNN accelerator with efficient parallelization methods for SR applications. First, we propose a new methodology for optimizing the deconvolutional neural networks (DCNNs) used for increasing feature maps. Secondly, we propose a novel method to optimize CNN dataflow so that the SR algorithm can be driven at low power in display applications. Finally, we quantize and compress a DCNN-based SR algorithm into an optimal model for efficient inference using on-chip memory. We present an energyefficient architecture for SR and validate our architecture on a mobile panel with quad-high-definition (QHD) resolution. Our experimental results show that, with the same hardware resources, the proposed DCNN accelerator achieves a throughput up to 108 times greater than that of a conventional DCNN accelerator. In addition, our SR system achieves an energy efficiency of 144.9 GOPS/W, 293.0 GOPS/W, and 500.2 GOPS/W at SR scale factors of 2, 3, and 4, respectively. Furthermore, we demonstrate that our system can restore HR images to a high quality while greatly reducing the data bit-width and the number of parameters compared to conventional SR algorithms.

No abstract

Metastatic solid tumors are aggressive and mostly drug resistant leading to few treatment options and poor prognosis as seen with clear cell renal cell carcinoma (ccRCC) and triple negative breast cancer (TNBC). Therefore the identification of new therapeutic regimes for the treatment of metastatic disease is desirable. ccRCC and TNBC cell lines were treated with the HDAC inhibitor romidepsin and the methyltransferase inhibitor decitabine, two epigenetic modifying drugs approved by the FDA for the treatment of various hematologic malignancies. Cell proliferation analysis, flow cytometry, quantitative PCR and immuno-blotting techniques were utilized to evaluate the antitumor synergy of this drug combination and identify the re-expression of epigenetically silenced tumor suppressor genes. Combinatorial treatment of metastatic TNBC and stage 4 ccRCC cell lines with romidepsin/decitabine leads to synergistic inhibition of cell growth and induction of apoptosis above levels of individual drug treatments alone. Synergistic re-expression of the tumor suppressor gene secreted frizzled-related protein one (sFRP1) was observed in combinatorial drug treated groups. Silencing sFRP1 (shRNA) prior to combinatorial drug treatment demonstrated that sFRP1 mediates the growth inhibitory and apoptotic activity of combined romidepsin/decitabine. Furthermore, addition of recombinant sFRP1 to ccRCC or TNBC cells inhibits cell growth in a dose-dependent manner through the induction of apoptosis identifying that epigenetic silencing of sFRP1 contributes to renal and breast cancer cell survival. Combinatorial treatment with romidepsin and decitabine in drug resistant tumors is a promising treatment strategy. Moreover, recombinant sFRP1 may be a novel therapeutic strategy for cancers with suppressed sFRP1 expression.