Search

Deep convolutional neural networks have been demonstrated to be effective for SISR in recent years. On the one hand, residual connections and dense connections have been used widely to ease forward information and backward gradient flows to boost performance. However, current methods use residual connections and dense connections separately in most network layers in a sub-optimal way. On the other hand, although various networks and methods have been designed to improve computation efficiency, save parameters, or utilize training data of multiple scale factors for each other to boost performance, it either do super-resolution in HR space to have a high computation cost or can not share parameters between models of different scale factors to save parameters and inference time. To tackle these challenges, we propose an efficient single image super-resolution network using dual path connections with multiple scale learning named as EMSRDPN. By introducing dual path connections inspired by Dual Path Networks into EMSRDPN, it uses residual connections and dense connections in an integrated way in most network layers. Dual path connections have the benefits of both reusing common features of residual connections and exploring new features of dense connections to learn a good representation for SISR. To utilize the feature correlation of multiple scale factors, EMSRDPN shares all network units in LR space between different scale factors to learn shared features and only uses a separate reconstruction unit for each scale factor, which can utilize training data of multiple scale factors to help each other to boost performance, meanwhile which can save parameters and support shared inference for multiple scale factors to improve efficiency. Experiments show EMSRDPN achieves better performance and comparable or even better parameter and inference efficiency over SOTA methods., Comment: 21 pages, 9 figures, 5 tables

NF1 inactivation occurs in specific human cancers, including juvenile myelomonocytic leukemia, an aggressive myeloproliferative disorder of childhood. However, evidence suggests that Nf1 loss alone does not cause leukemia.We therefore hypothesized that inactivation of the Nf1 tumor suppressor gene requires cooperating mutations to cause acute leukemia. To search for candidate genes that cooperate with Nf1 deficiency in leukemogenesis, we performed a forward genetic screen using retroviral insertion mutagenesis in Nf1 mutant mice. We identified 43 common proviral insertion sites that contain candidate genes involved in leukemogenesis. One of these genes, Bc111a, confers a growth advantage in cultured Nf1 mutant hematopoietic cells and causes early onset of leukemia of either myeloid or lymphoid lineage in mice when expressed in Nf1-deficient bone marrow. Bc111a-expressing cells display compromised p21Cip1 induction, suggesting that Bc111a's oncogenic effects are mediated, in part, through suppression of p21Cip1. Importantly, Bc111a is expressed in human chronic myelomonocytic leukemia and juvenile myelomonocytic leukemia samples. A subset of AML patients, who had poor outcomes, of 16 clusters, displayed high levels of BCL11A in leukemic cells. These findings suggest that deregulated Bc111a cooperates with Nf1 in leukemogenesis, and a therapeutic strategy targeting the BCL11A pathway may prove beneficial in the treatment of leukemia.