The field of machine learning has become an increasingly budding area of research as more efficient methods are needed in the quest to handle more complex image detection challenges. To solve the problems of agriculture is more and more important because food is the fundamental of life. However, the detection accuracy in recent corn field systems are still far away from the demands in practice due to a number of different weeds. This paper presents a model to handle the problem of corn leaf detection in given digital images collected from farm field. Based on results of experiments conducted with several state-of-the-art models adopted by CNN, a region-based method has been proposed as a faster and more accurate method of corn leaf detection. Being motivated with such unique attributes of ResNet, we combine it with region based network (such as faster rcnn), which is able to automatically detect corn leaf in heavy weeds occlusion. The method is evaluated on the dataset from farm and we make an annotation ourselves. Our proposed method achieves significantly outperform in corn detection system.
There has been increased use of self-expandable metal stents (SEMS) in treating malignant colorectal obstruction (MCO). The aim of this study was to investigate factors that are associated with the outcomes of SEMS placement for MCO.Clinical data from patients who underwent SEMS placement for MCO at 6 hospitals in Honam province of South Korea between 2009 and 2018 were reviewed retrospectively. Eight hundred two patients were identified and their data were analyzed. Technical success, clinical success, complications, and predictors of outcome were included as main outcome measures.Technical and clinical success rates were 98.8% (792/802) and 90.1% (723/802), respectively. Complications including stent migration, stent occlusion due to tumor ingrowth and outgrowth, perforation, bacteremia/fever, and bleeding occurred in 123 (15.3%) patients. In multivariate regression analyses, procedure time was significantly associated with the technical success of SEMS placement (P = .001). Longer length of obstruction, the use of covered stent, and longer procedure time were significant independent predictive factors for the clinical success of SEMS placement (odds ratio [OR] 0.974 (95% confidence interval [CI] 0.950-0.990); P = .043, OR 0.255 (95% CI 0.138-0.471); P < .001, and OR 0.957 (95% CI 0.931-0.984); P = .002, respectively). Stage IV colorectal cancer and the use of covered stent were significant independent predictive factors for the development of complications after SEMS placement (OR 2.428 (95% CI 1.407-4.188); P = .001 and OR 3.329 (95% CI 2.060-5.378); P < .001, respectively).Longer length of obstruction, the use of covered stent, and longer procedure time were associated with lower clinical success rates. Having stage IV colorectal cancer and the use of covered stents were associated with an increased risk of complications.
Abstract A recently rising question of the applicability of two-dimensional (2D) materials to membranes of enhanced performance in water technology is drawing attention increasingly. At the center of the attention lies graphene, an atom-thick 2D material, for its readiness and manufacturability. This review presents an overview of recent research activities focused on the fundamental mass transport phenomena of two feasible membrane architectures from graphene. If one could perforate pores in a pristine impermeable graphene sheet with dimensional accuracy, the perforated 2D orifice would show unrivaled permeation of gases and liquids due to the 0D atomic barrier. If possibly endowed with selectivity, the porous graphene orifice would avail potentially for membrane separation processes. For example, it is noteworthy that results of molecular dynamics simulations and several early experiments have exhibited the potential use of the ultrathin permeable graphene layer having sub-nanometer-sized pores for a water desalination membrane. The other membrane design is obtainable by random stacking of moderately oxidized graphene platelets. This lamellar architecture suggests the possibility of water treatment and desalination membranes because of subnanometric interlayer spacing between two adjacent graphene sheets. The unique structure and mass transport phenomena could enlist these graphene membrane architectures as extraordinary membrane material effective to various applications of membrane technology including water treatment. Graphic abstract
Systemic lupus erythematosus (SLE) is an autoimmune disease characterized by the production of diverse autoantibodies with various systemic organ involvements. In patients with SLE, autoantibodies, such as antinuclear antibody (ANA) and anti-dsDNA antibody, play an important role not only in diagnosing the disease, but also representing the pathogenesis of the disease. ANA is the main screening tool in diagnosis and serum complement levels and anti-dsDNA antibody level are closely related to the disease activities. Nevertheless, exceptionally, some patients represent with negative ANA and/or anti-dsDNA antibody leading to diffi-culties in diagnosing the disease. Here, we report a case of 37-year old female SLE patient with negative ANA, negative anti-dsDNA antibody, and positive anti-Ro/SSA antibody, which manifested with nephrotic syndrome.
Self-expandable metal stent (SEMS) placement is commonly used for palliation of left-sided malignant colorectal obstruction (MCO). However, right-sided MCO is usually treated surgically. Recent studies that compared palliative SEMS insertion and emergency surgery in right-sided MCOs have reported conflicting results. This study aimed to compare the effectiveness of palliative SEMS placement in left-sided MCOs and right-sided MCOs and to investigate the predictive factors for clinical success and risk factors for complications. Data from 469 patients who underwent palliative SEMS placement for MCO at 6 hospitals in the Honam province of South Korea between 2009 and 2018 were reviewed. Among them, 69 patients with right-sided MCO and 400 patients with left-sided MCO who underwent SEMS placement for palliative purposes were enrolled. Clinical success, overall survival, complications, and predictive factors for clinical success and risk factors for complications were included as the main outcome measures. The clinical success rates were 97.1% (65/67) in right-sided MCO patients and 88.2% (353/400) in left-sided MCO patients. Complications including stent migration, tumor ingrowth, outgrowth, perforation, bacteremia/fever, and bleeding occurred in 10.1% (7/69) of right-sided MCO patients and 19.9% (79/400) of left-sided MCO patients. The mean overall survival of right-sided MCO was 28.02 months and 18.23 months for left-sided MCO. In multivariate logistic regression analysis, T3 stage tumors and the use of uncovered stents were significant factors for the clinical success of SEMS. The use of covered stents and performance status score of 0 to 2 were independent significant risk factors for complications. Palliative SEMS placement in right-sided MCO showed better clinical success rates than left-sided MCO. The use of uncovered stents is recommended for higher clinical success rates and lower complication rates.
An attention guided convolutional neural network (CNN) for the classification of breast cancer histopathology images is proposed. Neural networks are generally applied as black box models and often the network's decisions are difficult to interpret. Making the decision process transparent, and hence reliable is important for a computer-assisted diagnosis (CAD) system. Moreover, it is crucial that the network's decision be based on histopathological features that are in agreement with a human expert. To this end, we propose to use additional region-level supervision for the classification of breast cancer histopathology images using CNN, where the regions of interest (RoI) are localized and used to guide the attention of the classification network simultaneously. The proposed supervised attention mechanism specifically activates neurons in diagnostically relevant regions while suppressing activations in irrelevant and noisy areas. The class activation maps generated by the proposed method correlate well with the expectations of an expert pathologist. Moreover, the proposed method surpasses the state-of-the-art on the BACH microscopy test dataset (part A) with a significant margin.
Weeds in agricultural farms are aggressive growers which compete for nutrition and other resources with the crop and reduce production. The increasing use of chemicals to control them has inadvertent consequences to the human health and the environment. In this work, a novel neural network training method combining semantic graphics for data annotation and an advanced encoder-decoder network for (a) automatic crop line detection and (b) weed (wild millet) detection in paddy fields is proposed. The detected crop lines act as a guiding line for an autonomous weeding robot for inter-row weeding, whereas the detection of weeds enables autonomous intra-row weeding. The proposed data annotation method, semantic graphics, is intuitive, and the desired targets can be annotated easily with minimal labor. Also, the proposed "extended skip network" is an improved deep convolutional encoder-decoder neural network for efficient learning of semantic graphics. Quantitative evaluations of the proposed method demonstrated an increment of 6.29% and 6.14% in mean intersection over union (mIoU), over the baseline network on the task of paddy line detection and wild millet detection, respectively. The proposed method also leads to a 3.56% increment in mIoU and a significantly higher recall compared to a popular bounding box-based object detection approach on the task of wild-millet detection.
A method for the automatic detection of diseases/infestations in paprika cultivation using AI is investigated. Powdery mildew was the paprika disease observed during hydroponic cultivation in a greenhouse environment. The two-spotted-spider-mite was the pest. Paprika leaves with either the disease or the pest were automatically detected using a Faster R-CNN network architecture. The detection performance was high with mAP 96.76 %. The training and testing arrangements and the training data set are described in detail.
'%3e%3cg id='b'%3e%3cpath id='a' stroke='%23000' stroke-width='2' d='M-6-25H6m-12 3H6m-12 3H6'/%3e%3cuse xlink:href='%23a' y='44'/%3e%3c/g%3e%3cpath stroke='%23fff' d='M0 17v10'/%3e%3ccircle r='12' fill='%23cd2e3a'/%3e%3cpath fill='%230047a0' d='M0-12A6 6 0 0 0 0 0a6 6 0 0 1 0 12 12 12 0 0 1 0-24z'/%3e%3c/g%3e%3cg transform='rotate(-123.69)'%3e%3cuse xlink:href='%23b'/%3e%3cpath stroke='%23fff' d='M0-23.5v3M0 17v3.5m0 3v3'/%3e%3c/g%3e%3c/svg%3e)