Search

Post-polypectomy syndrome (PPS) results from electrocoagulation injury to the bowel wall that induces a transmural burn and localized peritonitis. It has a good prognosis; however, there are exceptions when complications are observed. We here report a case of a 50-year-old man who developed lumbosacral pain and high fever with chills four days after colonoscopy, during which polypectomy was performed by endoscopic mucosal resection (EMR) and argon plasma coagulation (APC). Both the plain abdominal film and abdominal CT scan showed no free air, and lumbar CT showed no apparent lesions, which satisfied the diagnosis of PPS. However, the patient was in a critical condition as he developed septic shock caused by bacteremia. Following active treatment, the patient's condition rapidly improved. Therefore, we suggest that clinicians should consider the severity of PPS with sepsis and colon transmural burn. Patients with a diagnosis of PPS should be admitted to the hospital for observation and treatment to avoid adverse consequences.

Resistance to chemotherapy drugs, such as adriamycin (ADR), is a common problem in acute myeloid leukemia (AML) patients. We hypothesized that the natural compound resveratrol (Res) may reverse AML drug resistance through the PI3K/Akt/Nrf2 pathway. We investigated the in vitro effect of Res using human promyelocytic leukemia cells (HL-60) and the ADR-resistant cell line (HL-60/ADR) and treated with either Res or ADR + Res. Cellular proliferation inhibition rate, auto-fluorescence intensity of ADR in HL-60/ADR cells and HL-60 cells, mRNA expression of

Imperfect lightness conditions usually lower the visual quality of an image by bringing in unclear image details and poor image contrast. Traditional low-light enhancement models based on one single input are often limited in avoiding the effect of over-enhancement or under-enhancement. Models based on fusing multiple input sources usually perform well in relieving this issue, as they can harmonize the complementary visual appearances of a same scene provided by different sources. Nevertheless, these models still have difficulty in dealing with the situation that only one input is at hand, which usually happens in many practical situations. In this paper, we propose a low-light enhancement model that artificially enriches input sources and then seamlessly fuses them. Specifically, with an input image, we first generate multiple enhanced images based on a lightness-aware camera response model. These images are then fused at mid-level based on a patch-based image decomposition model. To validate our model, we conduct qualitative and quantitative comparisons with several state-of-the-art single-source and multi-source models on a collection of real-world images. Experimental results show that our model better improves the image quality in terms of visual naturalness and aesthetics.

This paper proposes a deep learning-based fast grasp detection method with a small dataset for robotic bin-picking. We consider the problem of grasping stacked up mechanical parts on a planar workspace using a parallel gripper. In this paper, we use a deep neural network to solve the problem with a single depth image. To reduce the computation time, we propose an edge-based algorithm to generate potential grasps. Then, a convolutional neural network (CNN) is applied to evaluate the robustness of all potential grasps for bin-picking. Finally, the proposed method ranks them and the object is grasped by using the grasp with the highest score. In bin-picking experiments, we evaluate the proposed method with a 7-DOF manipulator using textureless mechanical parts with complex shapes. The success ratio of grasping is 97%, and the average computation time of CNN inference is less than 0.23[s] on a laptop PC without a GPU array. In addition, we also confirm that the proposed method can be applied to unseen objects which are not included in the training dataset.

It has become very popular to take photographs in everyone’s daily life. However, the visual quality of a photograph is not always guaranteed due to various factors. One common factor is the low-light imaging condition, which conceals visual information and degenerates the quality of a photograph. It is preferable for a low-light image enhancement model to complete the following tasks: improving contrast, preserving details, and keeping robust to noise. To this end, we propose a simple but effective enhancing model based on the simplified Retinex theory, of which the key is to estimate a good illumination map. In our model, we apply an iterative self-guided filter to refine the initial estimation of an illumination map, making it aware of local structure of image contents. In experiments, we validate the effectiveness of our method in various aspects, and compare our model with several state-of-the-art ones. The results show that our method effectively adjusts the global image contrast, recovers the concealed details and keeps the robustness against noise.

To automate the bin picking task with robots, pose estimation is the key challenge which identifies and locates objects, thus the robot can pick and manipulate the object in an accurate and reliable way. This paper proposes a novel solution which combines a machine learning based 2D object localization and a non-machine learning based 3D pose estimation method to estimate the pose of randomly piled up industrial parts. Given an image of a scene, the target part is localized in 2D first and its result is then used to crop the point cloud of the target part. Using the cropped point cloud and Boundary-to-Boundary-using-Directional-Tangent-Line (B2B-DTL) point pair feature, a novel descriptor, the proposed method could estimate the pose of industrial parts whose point clouds lack key details, for example, the point cloud of ridges of a part. Our algorithm is evaluated against real scenes and its experimental results show that the proposed method is sufficiently accurate and its online computation time is short, which makes it could be used in the real factory environment.

The visual quality of photographs can be seriously affected by various low-light conditions. The image enhancement methods based on fusion achieve satisfying results. However, this poses challenges to the situation that there are no fusion sources and only a single low-light image is at hand. In this paper, we propose a single low-light enhancement method by fusing multiple sources, including the original image and its intermediate enhancements generated by a simplified Retinex model. We combine these sources at a mid-level image representation based on a patch-based decomposition model. Compared with other state-of-the-art methods, visual and quantitative results show that our method effectively improves visual effects in terms of lightness, color harmony and vividness.