Search

To develop a fully automated deep learning model for adrenal segmentation and to evaluate its performance in classifying adrenal hyperplasia.This retrospective study evaluated automated adrenal segmentation in 308 abdominal CT scans from 48 patients with adrenal hyperplasia and 260 patients with normal glands from 2010 to 2021 (mean age, 42 years; 156 women). The dataset was split into training, validation, and test sets at a ratio of 6:2:2. Contrast-enhanced CT images and manually drawn adrenal gland masks were used to develop a U-Net-based segmentation model. Predicted adrenal volumes were obtained by fivefold splitting of the dataset without overlapping the test set. Adrenal volumes and anthropometric parameters (height, weight, and sex) were utilized to develop an algorithm to classify adrenal hyperplasia, using multilayer perceptron, support vector classification, a random forest classifier, and a decision tree classifier. To measure the performance of the developed model, the dice coefficient and intraclass correlation coefficient (ICC) were used for segmentation, and area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity were used for classification.The model for segmenting adrenal glands achieved a Dice coefficient of 0.7009 for 308 cases and an ICC of 0.91 (95% CI, 0.90-0.93) for adrenal volume. The models for classifying hyperplasia had the following results: AUC, 0.98-0.99; accuracy, 0.948-0.961; sensitivity, 0.750-0.813; and specificity, 0.973-1.000.The proposed segmentation algorithm can accurately segment the adrenal glands on CT scans and may help clinicians identify possible cases of adrenal hyperplasia.• A deep learning segmentation method can accurately segment the adrenal gland, which is a small organ, on CT scans. • The machine learning algorithm to classify adrenal hyperplasia using adrenal volume and anthropometric parameters (height, weight, and sex) showed good performance. • The proposed segmentation algorithm may help clinicians identify possible cases of adrenal hyperplasia.

To investigate the optimal input matrix size for deep learning-based computer-aided detection (CAD) of nodules and masses on chest radiographs. We retrospectively collected 2088 abnormal (nodule/mass) and 352 normal chest radiographs from two institutions. Three thoracic radiologists drew 2758 abnormalities regions. A total of 1736 abnormal chest radiographs were used for training and tuning convolutional neural networks (CNNs). The remaining 352 abnormal and 352 normal chest radiographs were used as a test set. Two CNNs (Mask R-CNN and RetinaNet) were selected to validate the effects of the squared different matrix size of chest radiograph (256, 448, 896, 1344, and 1792). For comparison, figure of merit (FOM) of jackknife free-response receiver operating curve and sensitivity were obtained. In Mask R-CNN, matrix size 896 and 1344 achieved significantly higher FOM (0.869 and 0.856, respectively) for detecting abnormalities than 256, 448, and 1792 (0.667–0.820) (p

Scanning electron microscopy (SEM) is a high-resolution imaging technique with subnanometer spatial resolution that is widely used in materials science, basic science, and nanofabrication. However, conducting SEM is rather complex due to the nature of using an electron beam and the many parameters that must be adjusted to acquire high-quality images. Only trained operators can use SEM equipment properly, meaning that the use of SEM is restricted. To broaden the usability of SEM, we propose an autofocus method for a SEM system based on a dual deep learning network, which consists of an autofocusing-evaluation network (AENet) and an autofocusing-control network (ACNet). The AENet was designed to evaluate the quality of given images, with scores ranging from 0 to 9 regardless of the magnification. The ACNet can delicately control the focus of SEM online based on the AENet’s outputs for any lateral sample position and magnification. The results of these dual networks showed successful autofocus performance on three trained samples. Moreover, the robustness of the proposed method was demonstrated by autofocusing on unseen samples. We expect that our autofocusing system will not only contribute to expanding the versatility of SEM but will also be applicable to various microscopes.

In this study, K(Ta0.6Nb0.4)O3 thin films were fabricated by sol–gel method and their structural and electrical properties were measured to investigate the applicability for electrocaloric devices. As the thickness of thin films increased, crystallinity was improved. The average thickness of the thin films with a single coating was about 130–140 nm. Dielectric constant, dielectric loss, remanent polarization, and coercive field of the 6 times coated thin KTN films at 30 °C were 4920, 0.492, 18.49 μC/cm2 and 54.7 kV/cm, respectively. As the number of coatings increased, the temperature at which the decrease in remanent polarization began also subsided. Likewise, the rate of change in remanent polarization increased as the temperature became higher. When a voltage of 220 kV/cm was applied to the 3 times coated KTN films, the electrocaloric property was 2.95 °C. When the 3 times coated thin KTN films were at 75 °C, electrocaloric property for a unit electric field was about 1.47 °C.

In this study, welding-induced out-of-plane distortion of stainless steel sheets is examined in the case of applying air-atomized mist jet on the rear side of the weld pool. An internally mixed type air atomizing spray nozzle is attached to the welding torch and both are moved at the same speed and direction onto the welding centerline. The test specimens are 2 mm thick grade 304 stainless steel plates. Bead-on-plate welds are made on the plates using a DC Gas tungsten arc welding (GTAW) process. The measured welding distortion and residual stresses are compared with those of a conventional GTAW welding process. The experimental results show that the welds performed with a trailing heat sink using air-atomized mist have almost zero out-of-plane distortion. In addition, a slight welding residual stress reduction is achieved.

Recently, Car makers have required a light weight of vehicles in relation to fuel efficiency. In that respect, dissimilar welding method is very important in the car body assembly process. In this study, resistance spot welding was performed to evaluate the welding quality and mechanical properties for dissimilar materials of the cold rolled DP590 and Al 5052. The welding experiment was carried out changing with the welding current and the other process parameters such as electrode force, weld time, squeezing time and holding time were fixed with the basic welding condition of the ISO18278-2 standard. Especially, the shear tensile strength was evaluated at each condition and the cross-sectional observation for weld zone such as nugget diameter was performed. As a result, the thickness of IMC (Intermetallic Compound) layer at 11.5 kA was thinner than those of 9.5 kA and 10.5 kA conditions. On the other hand, Delta-spot welding experiment was performed using various process tapes at welding current of 11.5 kA. The other conditions were same with that of the conventional spot welding. Sound welds was formed by applying to the optimum welding conditions in the Delta-spot process due to the thermal balance of dissimilar materials.

To investigate the reproducibility of computer-aided detection (CAD) for detection of pulmonary nodules and masses for consecutive chest radiographies (CXRs) of the same patient within a short-term period. A total of 944 CXRs (Chest PA) with nodules and masses, recorded between January 2010 and November 2016 at the Asan Medical Center, were obtained. In all, 1092 regions of interest for the nodules and mass were delineated using an in-house software. All CXRs were randomly split into 6:2:2 sets for training, development, and validation. Furthermore, paired follow-up CXRs (n = 121) acquired within one week in the validation set, in which expert thoracic radiologists confirmed no changes, were used to evaluate the reproducibility of CAD by two radiologists (R1 and R2). The reproducibility comparison of four different convolutional neural net algorithms and two chest radiologists (with 13- and 14-years’ experience) was conducted. Model performances were evaluated by figure-of-merit (FOM) analysis of the jackknife free-response receiver operating curve and reproducibility rates were evaluated in terms of percent positive agreement (PPA) and Chamberlain’s percent positive agreement (CPPA). Reproducibility analysis of the four CADs and R1 and R2 showed variations in the PPA and CPPA. Model performance of YOLO (You Only Look Once) v2 based eDenseYOLO showed a higher FOM (0.89; 0.85–0.93) than RetinaNet (0.89; 0.85–0.93) and atrous spatial pyramid pooling U-Net (0.85; 0.80–0.89). eDenseYOLO showed higher PPAs (97.87%) and CPPAs (95.80%) than Mask R-CNN, RetinaNet, ASSP U-Net, R1, and R2 (PPA: 96.52%, 94.23%, 95.04%, 96.55%, and 94.98%; CPPA: 93.18%, 89.09%, 90.57%, 93.33%, and 90.43%). There were moderate variations in the reproducibility of CAD with different algorithms, which likely indicates that measurement of reproducibility is necessary for evaluating CAD performance in actual clinical environments.

The automotive industry has a target goal to improve fuel consumption due to restricted exhaust gas regulation. For this reason, the applicability of lightweight material, Al alloys, Mg alloys is also being expanded. In this concept, high strength steel, DP590 and light alloy, AL5052 are joined in the right place of the car body. However, it is difficult to join to steel and aluminum by conventional fusion welding. Generally, in respect to dissimilar metal joining by fusion welding, intermetallic compound layer is formed at the joint interface, hot cracking is generated. In this study, the effect of the current waveform on the mechanical characteristics and microstructure in Delta spot welding process of dissimilar metal was investigated. As results, Intermetallic compound (IMC) layer was reduced from 2.355 μm to 1.09 μm by using Delta spot welding process; also the welding current range improved by 50% in the delta spot welding, higher than in the inverter resistance welding. To conclude, the delta spot welding process adopting the process tapes contributes to improving the welding quality for dissimilar metals (Al5052 and DP590) due to a decrease in IMC layer.

Background: The rate of metabolic syndrome (MetS) is exceptionally high in Korea, and the health risks and dietary implications of MetS have been reported extensively. Although most meals include combination foods, little is known about metabolic responses to meals containing various foods (such as a hamburger, which typically includes meat, bread, and vegetables). The purpose of this study was to compare glycemic responses and postprandial-triglyceride (PTG) concentrations after consuming each of four test meals: bibimbap, kimbap, pork cutlet, and hamburger. Methods: A clinical trial was carried out with 32 men using an open 4 × 4 × 4 treatment, sequence, and period crossover design. The blood-glucose concentrations, incremental area under the curve (iAUC), glycemic index (GI), insulinogenic index, and PTG after each test meal were determined. Results: The iAUC2h and Cmax responses were proportional to the amount of carbohydrates in the experimental diets [i.e., higher (p

The effects of ultrathin fibers using electrospun m-aramid nanofibrous membranes were evaluated to determine their performance as filters for water purification. The M150 membrane (average fiber diameter: 150 nm) was higher air-permeability than the M120 membrane (average fiber diameter: 120 nm). The commercial GSWP filter had much lower air-permeability than the electrospun membranes. The mean flow pore diameters of the electrospun membranes were much larger than the GSWP due to their high air permeabilities. Increased membrane thickness resulted in decreased airpermeability, mean pore size, and largest detected pore size. The M150 and M120 membranes with thickness of 10 μm showed approximately 5 times higher water flux than the GSWP filter. However, these membranes surprisingly showed higher particle rejections (99.32% and 99.62%, respectively) in the filtration of 0.2-μm PS latex particles compared with the GSWP filter (95.91%). This result indicates that highly branched, ultrafine fibrous structures of the m-aramid nanofibrous membranes contribute to the great increase in the particle rejection efficiency without loss of water flux (or airpermeability). From these results, the electrospun m-aramid nanofibrous membranes are expected to be used as a novel microfilter with a super high flux and high rejection efficiency.

In the automotive industry, vehicle weight reduction techniques have been actively studied to improve the rate of fuel consumption and to cope with the regulation restricting exhaust gas. For this reason, advanced high-strength steel (AHSS) is preferred in the automobile industry as its tensile strength is 590 MPa and over. In this study, to obtain the optimum welding condition, the friction stir welding (FSW) process applied to AHSS was considered. The FSW experiment was performed on a stir plate using a Si3N4 tool and a 1.4-mm-thick DP590 steel sheet manufactured by cold rolling. In addition, to investigate the temperature distribution of the advancing and retreating sides in the welding state, the tool rotation speed of 800 rpm, and the welding speed of 180 mm/min, a K-type thermocouple was inserted in the backing plate, and the peak temperature was evaluated at each point. Especially, the correlation between the heat input per unit length and the formation of the FSW zone was minutely analyzed.

It is important to overlap the even bead continuously to automate the overlay welding process. To obtain the even surface bead on overlay welded product, optimization of pitch distance between the each pass is one of the methods. Therefore, this study determines the optimal second pass pitch distance with respect to welding parameter such as torch position and welding speed. Variety of welding experiments was carried out with different welding parameters using Central Composite Rotatable Design (CCRD) method. Based on the experimental results, empirical models were developed after analyzing correlation between welding parameters and bead height, width and decentration of overlay welding. Finally, optimized empirical models were developed for predicting bead height and width by analyzing the main effect of each factor which affects bead geometry and their influence on interaction. In addition, a mathematical model was developed which determines the bead profile for visualizing the bead shapes using the Matlab curve fitting tool. These two models were combined for one of the overlay bead shape model. Furthermore, the mathematical model is then used to determine the second pass pitch distance, by allowing the same area above and below the point of interaction between two overlay weld pass.

To investigate the clinical features in adult patients with febrile urinary tract infection (UTI) who visited the emergency department (ED) and to determine the predictive factors of bacteremia among the initial presenting clinical features.This retrospective cohort study was conducted at the ED of a tertiary hospital in Korea from 1 January 2012 to 31 December 2012. All adult patients who were diagnosed with febrile UTI and for whom data on blood and urine cultures were available were included in the study. Clinical examinations and laboratory tests were performed at the initial presentation.A total of the 325 patients with febrile UTI (median age: 60 years) were included for analysis, of whom 82 % were female. Bacteremia was detected in 106 of the 325 patients (32.6 %), with Escherichia coli the most frequent pathogen detected (59.7 % of cases). Between the bacteremic and non-bacteremic groups, there was significant difference in age (67 vs. 57 years, respectively), flank pain (16 vs. 7.8 %), suprapubic discomfort (0 vs. 4.6 %), body temperature (38.8 vs. 38.3 °C), respiratory rate (21 vs. 20/min), platelet count (170 vs. 186 × 10(3)/μL), C-reactive protein (10.2 vs. 8.3 mg/dL), and procalcitonin (1.5 vs. 0.3 ng/mL) (P0.05 for all). In the multivariate logistic regression analysis, age [odds ratio (OR) 1.03; 95 % confidence interval (CI) 1.01-1.05], systolic blood pressure of90 mmHg (OR 3.27; 95 % CI 1.13-9.45), body temperature of39 °C (OR 4.26; 95 % CI 2.28-7.96), and procalcitonin level of0.5 ng/dL (OR 2.03; 95 % CI 1.07-3.86) were significantly associated with bacteremia.Among our adult patients with febrile UTI, age, systolic blood pressure, body temperature, and procalcitonin were significantly associated with bacteremia. We therefore suggest that these factors should be considered when deciding upon treatment options for febrile UTI patients at the ED.

Selective reduction of NO by CH4 in the presence of excess O2 has been studied on cobalt-ion-exchanged synthetic clinoptilolite (Co-CLI) zeolite. The catalytic results are compared with those obtained from Co-ZSM-5 and Co-FER, the two most widely studied catalysts for this reaction. At T ≥ 500 oC, Co-CLI is much more active for NO reduction than Co-ZSM-5, but it is less active than Co-FER. However, it is found that the selectivity for CH4 toward reacting with NO at high temperatures (T ≥ 450 oC) is higher on Co-CLI than on Co-ZSM-5 or Co-FER. On the basic of the overall catalytic results, the importance of zeolite pore size as a structural parameter influencing the activity of intrazeolitic Co2+ ions for NO reduction by CH4 is discussed.

A xylanase was purified from a commercial crude xylanase, Pulpzyme HC, and used for the bleaching of kraft pulp in the absence or in the presence of nonionic surfactants, Tween 20, Tween 80, and Igepal C930. The purified xylanase has a molecular weight of 23,500 as determined by a reducing SDS-PAGE. Tween 20 was most effective to enhance the efficiency of the enzymatic bleaching of kraft pulp by xylanase.

In the present study a shell contact problem with geometric and material nonlinearities is efficiently formulated by utilizing the mathematical programming method. The contact surface is assumed unbonded and frictionless. An incremental analysis by the updated Lagrangian approach is used. Two representative problems are treated to show modeling of the shell contact and the proposed solution method. The results are compared with existing solutions and those calculated by a commerical package.