How can we model node representations to accurately infer the signs of missing edges in a signed social graph? Signed social graphs have attracted considerable attention to model trust relationships ...between people. Various representation learning methods such as network embedding and graph convolutional network (GCN) have been proposed to analyze signed graphs. However, existing network embedding models are not end-to-end for a specific task, and GCN-based models exhibit a performance degradation issue when their depth increases. In this paper, we propose Signed Diffusion Network (SidNet), a novel graph neural network that achieves end-to-end node representation learning for link sign prediction in signed social graphs. We propose a new random walk based feature aggregation, which is specially designed for signed graphs, so that SidNet effectively diffuses hidden node features and uses more information from neighboring nodes. Through extensive experiments, we show that SidNet significantly outperforms state-of-the-art models in terms of link sign prediction accuracy.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Given a real-world graph, how can we measure relevance scores for ranking and link prediction? Random walk with restart (RWR) provides an excellent measure for this and has been applied to various ...applications such as friend recommendation, community detection, anomaly detection, etc. However, RWR suffers from two problems: 1) using the same restart probability for all the nodes limits the expressiveness of random walk, and 2) the restart probability needs to be manually chosen for each application without theoretical justification. We have two main contributions in this paper. First, we propose Random Walk with Extended Restart (RWER), a random walk based measure which improves the expressiveness of random walks by using a distinct restart probability for each node. The improved expressiveness leads to superior accuracy for ranking and link prediction. Second, we propose SuRe (Supervised Restart for RWER), an algorithm for learning the restart probabilities of RWER from a given graph. SuRe eliminates the need to heuristically and manually select the restart parameter for RWER. Extensive experiments show that our proposed method provides the best performance for ranking and link prediction tasks.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Given a trained deep graph convolution network (GCN), how can we effectively compress it into a compact network without significant loss of accuracy? Compressing a trained deep GCN into a compact GCN ...is of great importance for implementing the model to environments such as mobile or embedded systems, which have limited computing resources. However, previous works for compressing deep GCNs do not consider the multi-hop aggregation of the deep GCNs, though it is the main purpose for their multiple GCN layers. In this work, we propose MustaD (Multi-staged knowledge Distillation), a novel approach for compressing deep GCNs to single-layered GCNs through multi-staged knowledge distillation (KD). MustaD distills the knowledge of 1) the aggregation from multiple GCN layers as well as 2) task prediction while preserving the multi-hop feature aggregation of deep GCNs by a single effective layer. Extensive experiments on four real-world datasets show that MustaD provides the state-of-the-art performance compared to other KD based methods. Specifically, MustaD presents up to 4.21%p improvement of accuracy compared to the second-best KD models.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Segmentation of normal organs is a critical and time-consuming process in radiotherapy. Auto-segmentation of abdominal organs has been made possible by the advent of the convolutional neural network. ...We utilized the U-Net, a 3D-patch-based convolutional neural network, and added graph-cut algorithm-based post-processing. The inputs were 3D-patch-based CT images consisting of 64 × 64 × 64 voxels designed to produce 3D multi-label semantic images representing the liver, stomach, duodenum, and right/left kidneys. The datasets for training, validating, and testing consisted of 80, 20, and 20 CT simulation scans, respectively. For accuracy assessment, the predicted structures were compared with those produced from the atlas-based method and inter-observer segmentation using the Dice similarity coefficient, Hausdorff distance, and mean surface distance. The efficiency was quantified by measuring the time elapsed for segmentation with or without automation using the U-Net. The U-Net-based auto-segmentation outperformed the atlas-based auto-segmentation in all abdominal structures, and showed comparable results to the inter-observer segmentations especially for liver and kidney. The average segmentation time without automation was 22.6 minutes, which was reduced to 7.1 minutes with automation using the U-Net. Our proposed auto-segmentation framework using the 3D-patch-based U-Net for abdominal multi-organs demonstrated potential clinical usefulness in terms of accuracy and time-efficiency.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Signal transducer and activator of transcription 3 (STAT3) is persistently activated in squamous cell carcinoma of the head and neck (SCCHN) and can cause uncontrolled cellular proliferation and ...division.
Thus, its targeted abrogation could be an effective strategy to reduce the risk of SCCHN. Resveratrol is known for its anti-cancer efficacy in a variety of cancer models.
The effect resveratrol on STAT3 activation, associated protein kinases, phosphatases, cellular proliferation and apoptosis was investigated.
We evaluated the effect of resveratrol on STAT3 signaling cascade and its regulated functional responses in SCCHN cells.
We found that HN3 and FaDu cells expressed strongly phosphorylated STAT3 on both tyrosine 705 and serine 727 residues as compared to other SCCHN cells. The phosphorylation was completely suppressed by resveratrol in FaDu cells, but not substantially in HN3 cells. STAT3 suppression was mediated through the inhibition of activation of upstream JAK2, but not of JAK1 and Src kinases. Treatment with the protein tyrosine phosphatase (PTP) inhibitor pervanadate reversed the resveratrol-induced down-regulation of STAT3, thereby indicating a critical role for a PTP. We also found that resveratrol induced the expression of the SOCS-1 protein and mRNA. Further, deletion of SOCS-1 gene by siRNA suppressed the induction of SOCS-1, and reversed the inhibition of STAT3 activation. Resveratrol down-regulated various STAT3-regulated gene products, inhibited proliferation, invasion, as well as induced the cell accumulation in the sub-G1 phase and caused apoptosis. Beside, this phytoalexin also exhibited the enhancement of apoptosis when combined with ionizing radiation treatment.
Our results suggest that resveratrol blocks STAT3 signaling pathway through induction of SOCS-1, thus attenuating STAT3 phosphorylation and proliferation in SCCHN cells.
Display omitted
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK, ZRSKP
To investigate the accuracy of the CyberKnife Xsight Lung Tracking System (XLTS) compared with that of a fiducial-based target tracking system (FTTS) using patient-specific lung phantoms.
...Three-dimensional printing technology was used to make individualized lung phantoms that closely mimicked the lung anatomy of actual patients. Based on planning computed tomographic data from 6 lung cancer patients who underwent stereotactic ablative radiation therapy using the CyberKnife, the volume above a certain Hounsfield unit (HU) was assigned as the structure to be filled uniformly with polylactic acid material by a 3-dimensional printer (3D Edison, Lokit, Korea). We evaluated the discrepancies between the measured and modeled target positions, representing the total tracking error, using 3 log files that were generated during each treatment for both the FTTS and the XLTS. We also analyzed the γ index between the film dose measured under the FTTS and XLTS.
The overall mean values and standard deviations of total tracking errors for the FTTS were 0.36 ± 0.39 mm, 0.15 ± 0.64 mm, and 0.15 ± 0.62 mm for the craniocaudal (CC), left-right (LR), and anteroposterior (AP) components, respectively. Those for the XLTS were 0.38 ± 0.54 mm, 0.13 ± 0.18 mm, and 0.14 ± 0.37 mm for the CC, LR, and AP components, respectively. The average of γ passing rates was 100% for the criteria of 3%, 3 mm; 99.6% for the criteria of 2%, 2 mm; and 86.8% for the criteria of 1%, 1 mm.
The XLTS has segmentation accuracy comparable with that of the FTTS and small total tracking errors.
Full text
Available for:
GEOZS, IJS, NUK, OILJ, UL, UM, UPUK
Struvite crystallization can recover nitrogen and phosphorus simultaneously from various kinds of wastewaters as a slow-release fertilizer. However, the enhancement of the removal efficiency of NH4-N ...is challenging because the molar concentration of NH4-N is higher than that of PO4-P in many types of sewage including digested sludge filtrate. In this study, phosphorus eluate was recovered from sewage sludge incineration ash (SSA) and applied to the struvite crystallization process to increase the removal efficiency of NH4-N for the digested sludge filtrate. Under acidic conditions, a maximum of 98.4% of phosphorus was eluted from SSA; in alkaline conditions, a maximum of 51.2% was eluted; and in sequential elution conditions with (NaOH+H2SO4), a maximum of 98.0% was eluted. Jar tests were performed by injecting three types of eluates (H2SO4 1 N_elulate, NaOH 1 N_elulate, and (NaOH+H2SO4)_eluate), and PO4-P was stably removed (>86%) under all tested conditions. When the NaOH 1 N_eluate was injected, the NH4-N removal efficiency was highest at 84.4%, followed by 78.4% with the (NaOH+H2SO4)_eluate, and 58.7% with the H2SO4 1 N_eluate at the molar ratio of Mg:P:N of 1.5:1.5:1. In addition, the sequential jar tests were conducted by injecting both the NaOH 1 N_eluate and (NaOH+H2SO4)_eluate. In the pH range of 8.5–9.5, the PO4-P and NH4-N removal efficiencies reached 92.3–94.5% and 97.9–99.1%, respectively. X-ray diffraction analyses confirmed that the majority of the crystal phases were struvite forms. Therefore, the combined application of both the NaOH 1 N_eluate and (NaOH+H2SO4)_eluate was adequate to enhance not only the phosphorus recovery but also the removal efficiencies of PO4-P and NH4-N. SSA recovering PO4-P could be utilized as a new phosphorus source in the struvite crystallization process.
The pyrolysis of food waste has high economic potential and produces several value-added products, such as gas, bio-oil, and biochar. In South Korea, biochar production from food waste is prohibited, ...because dioxins are generated during combustion caused by the chloride ions arising from the high salt content. This study is the first to examine the water quality and the applicability of food waste-based biochar as solid refuse fuel (SRF) based on a demineralization process. The calorific value increased after demineralization due to the removal of ionic substances and the high carbon content. The chloride ion removal rate after demineralization increased with the increasing pyrolysis temperature. A proximate analysis of biochar indicated that the volatile matter decreased, while ash and fixed carbon increased, with increasing pyrolysis temperature. At 300 °C pyrolysis temperature, all domestic bio-SRF standards were met. The organic matter concentration in water decreased with increasing carbonization temperature, and the concentrations of soluble harmful substances, such as volatile organic compounds (VOCs), were within the standards or non-detectable. These results suggest that biochar can be efficiently generated from food waste while meeting the emission standards for chloride ions, dissolved VOCs, ash, and carbon.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Background/Aims: Stereotactic body radiation therapy (SBRT) is used as an alternative ablative treatment in patients with hepatocellular carcinoma (HCC) not suitable for curative treatments. The ...purpose of this prospective study was to evaluate the long-term efficacy of SBRT for small (≤5 cm) HCCs.Methods: A phase II, single-arm clinical trial on SBRT for small HCCs was conducted at an academic tertiary care center. The planned SBRT dose was 45 Gy with a fraction size of 15-Gy over 3 consecutive days. The primary endpoint was 2-year local control rate. Radiologic responses were assessed according to the Response Evaluation Criteria in Solid Tumors (RECIST, version 1.1) and the modified RECIST criteria.Results: Between 2013 and 2016, 50 patients (53 lesions) were enrolled, with a median follow-up period of 47.8 months (range, 2.9–70.6). Patients’ age ranged from 41 to 74 years, and 80% were male. Median tumor size was 1.3 cm (range, 0.7–3.1). The 2- and 5-year local control rates were 100% and 97.1%, respectively. The 5-year overall survival rate was 77.6%. Six months after SBRT, radiologic responses were evident in 44 lesions (83%) according to the RECIST criteria and 49 (92.4%) according to the modified RECIST criteria. None of the patients showed grade ≥3 adverse events.Conclusions: SBRT showed excellent results as an ablative treatment for patients with small HCCs while showing minimal toxicities. SBRT can be a good alternative for both curative and salvage intents in patients with HCCs that are unsuitable for curative treatments.
To study the performance of a proposed deep learning-based autocontouring system in delineating organs at risk (OARs) in breast radiotherapy with a group of experts.
Eleven experts from two ...institutions delineated nine OARs in 10 cases of adjuvant radiotherapy after breast-conserving surgery. Autocontours were then provided to the experts for correction. Overall, 110 manual contours, 110 corrected autocontours, and 10 autocontours of each type of OAR were analyzed. The Dice similarity coefficient (DSC) and Hausdorff distance (HD) were used to compare the degree of agreement between the best manual contour (chosen by an independent expert committee) and each autocontour, corrected autocontour, and manual contour. Higher DSCs and lower HDs indicated a better geometric overlap. The amount of time reduction using the autocontouring system was examined. User satisfaction was evaluated using a survey.
Manual contours, corrected autocontours, and autocontours had a similar accuracy in the average DSC value (0.88 vs. 0.90 vs. 0.90). The accuracy of autocontours ranked the second place, based on DSCs, and the first place, based on HDs among the manual contours. Interphysician variations among the experts were reduced in corrected autocontours, compared to variations in manual contours (DSC: 0.89-0.90 vs. 0.87-0.90; HD: 4.3-5.8 mm vs. 5.3-7.6 mm). Among the manual delineations, the breast contours had the largest variations, which improved most significantly with the autocontouring system. The total mean times for nine OARs were 37 min for manual contours and 6 min for corrected autocontours. The results of the survey revealed good user satisfaction.
The autocontouring system had a similar performance in OARs as that of the experts' manual contouring. This system can be valuable in improving the quality of breast radiotherapy and reducing interphysician variability in clinical practice.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK