Analysis of the Region of Interest According to CNN Structure in Hierarchical Pattern Surface Inspection Using CAM.

A convolutional neural network (CNN), which exhibits excellent performance in solving image-based problem, has been widely applied to various industrial problems. In general, the CNN model was applied to defect inspection on the surface of raw materials or final products, and its accuracy also showed better performance compared to human inspection. However, surfaces with heterogeneous and complex backgrounds have difficulties in separating defects region from the background, which is a typical challenge in this field. In this study, the CNN model was applied to detect surface defects on a hierarchical patterned surface, one of the representative complex background surfaces. In order to optimize the CNN structure, the change in inspection performance was analyzed according to the number of layers and kernel size of the model using evaluation metrics. In addition, the change of the CNN's decision criteria according to the change of the model structure was analyzed using a class activation map (CAM) technique, which can highlight the most important region recognized by the CNN in performing classification. As a result, we were able to accurately understand the classification manner of the CNN for the hierarchical pattern surface, and an accuracy of 93.7% was achieved using the optimized model.

Estimating the phase volume fraction of multi-phase steel via unsupervised deep learning.

Advanced high strength steel (AHSS) is a steel of multi-phase microstructure that is processed under several conditions to meet the current high-performance requirements from the industry. Deep neural network (DNN) has emerged as a promising tool in materials science for the task of estimating the phase volume fraction of these steels. Despite its advantages, one of its major drawbacks is its requirement of a sufficient amount of training data with correct labels to the network. This often comes as a challenge in many areas where obtaining data and labeling it is extremely labor-intensive. To overcome this challenge, an unsupervised way of learning DNN, which does not require any manual labeling, is proposed. Information maximizing generative adversarial network (InfoGAN) is used to learn the underlying probability distribution of each phase and generate realistic sample points with class labels. Then, the generated data is used for training an MLP classifier, which in turn predicts the labels for the original dataset. The result shows a mean relative error of 4.53% at most, while it can be as low as 0.73%, which implies the estimated phase fraction closely matches the true phase fraction. This presents the high feasibility of using the proposed methodology for fast and precise estimation of phase volume fraction in both industry and academia.

Radiation exposure and fluoroscopically-guided interventional procedures among orthopedic surgeons in South Korea.

BACKGROUND: The use of fluoroscopically-guided interventional (FGI) procedures by orthopedic surgeons has been increasing. This study aimed to investigate the occupational radiation exposure among orthopedic surgeons in South Korea. METHODS: A nationwide survey of orthopedic surgeons was conducted in South Korea in October 2017. The dosimetry data of the participants were obtained from the National Dosimetry Registry. The orthopedic surgeons were categorized by job specialty [spine or trauma specialists, other orthopedic specialists, and residents], and descriptive statistics for the demographics and work-related characteristics were presented. Multivariable logistic regression analysis was used to evaluate the risk factors for the orthopedic surgeons who were not linked with the dosimetry data. RESULTS: Among the total participants (n = 513), 40.5% of the orthopedic surgeons spent more than 50% of their time working with the FGI procedures when compared with their overall work. The average frequency of the FGI procedures among the orthopedic surgeons was 12.3 days per month. Less than 30% of the participants were regularly provided with radiation monitoring badges. The proportion of subjects who always wore lead aprons and thyroid shields were 52 and 29%, respectively. The residents group experienced more unfavorable working conditions of radiation exposure than the other specialists. The dosimetry data were not significantly linked among the residents (odds ratio [OR] 2.10, 95% confidence interval [CI] 1.11-3.95) and orthopedic surgeons working at small hospitals (OR 4.76, 95% CI 1.05-21.51). CONCLUSIONS: Although orthopedic surgeons often performed FGI procedures, they wore protective gear less frequently, and a large proportion of orthopedic surgeons were not monitored by the national radiation dosimetry system. As the number of radiation procedures performed by the orthopedic surgeons increases, more intensive approaches are needed to reduce radiation exposure, especially for spine and trauma surgeons.

Association between depression and cardiovascular disease risk in general population of Korea: results from the Korea National Health and Nutrition Examination Survey, 2016.

BACKGROUND: Depression is considered as a risk factor for cardiovascular disease (CVD) and associated with changes in individuals' health status that might influence CVD risk. However, most studies have scrutinized this relationship on a rather narrower and specific study population. By focusing on general population of Korea, we sought to inspect the association of depression with CVD risk and cardiovascular risk factors. METHODS: The data from the first year (2016) of the 7th Korea National Health and Nutrition Examination Survey was used. Participants were classified by the Patient Health Questionnaire-9 (PHQ-9) score as such: normal group (PHQ-9 score 0-4), mild depression (MD) group (PHQ-9 score 5-9) and moderate and severe depression (MSD) group (PHQ-9 score 10-27). General linear model was used to analyze differences and the trend of mean CVD risk according to depression level. Adjusted odds ratios (AORs) were calculated by logistic regression to identify the association between depression and cardiovascular risk factors after adjusting for age. RESULTS: Mean CVD risk of MSD group was higher than that of normal group (p < 0.05). There was a tendency of CVD risk to increase as depression worsened (p < 0.01). Among men, MSD group was associated with current smoking (AOR, 2.97; 95% confidence interval [CI], 1.78-4.97), taking antihypertensive medications (AOR, 2.42; 95% CI, 1.26-4.66), increased fasting blood sugar (> 125 mg/dL; AOR, 2.37; 95% CI, 1.25-4.50) and taking diabetes medications (AOR, 3.08; 95% CI, 1.65-5.72). MD group was associated with current smoking (AOR, 1.60; 95% CI, 1.18-2.17). Among women, MSD group was associated with high body mass index (≥ 25 kg/m2; AOR, 1.61; 95% CI, 1.11-2.32), large waist circumference (≥ 85 cm; AOR, 1.63; 95% CI, 1.12-2.37), current smoking (AOR, 5.11; 95% CI, 3.07-8.52) and taking diabetes medications (AOR, 2.62; 95% CI, 1.68-4.08). MD group was associated with current smoking (AOR, 1.86; 95% CI, 1.18-2.93). CONCLUSIONS: We suggest that depression is associated with increased risk for CVD occurrence in general population of Korea.

Effect of Quenching Tempering-Post Weld Heat Treatment on the Microstructure and Mechanical Properties of Laser-Arc Hybrid-Welded Boron Steel.

In the present study, we have investigated the effect of post-welding heat treatment (PWHT) of quenching and tempering (QT) on the microstructure and mechanical properties of welded boron steel joints processed using laser-arc hybrid welding on two commercial filler materials, SM80 (Type-I) and ZH120 (Type-II). The microstructure and mechanical properties of the weld joints were characterized via optical microscopy, Vickers microhardness, and the uniaxial tensile test. The macrostructure of the weld joint was composed of a fusion zone (FZ), heat-affected zone (HAZ), and base metal zone (BMZ). After the QT-PWHT, the QT specimens revealed the V-shape hardness distribution across the weld joint, while the as-welded specimen exhibited the M-shape hardness distribution. As a result, the QT specimens revealed the premature fracture with little reduction in the area at the interface between the HAZ and FZ, while the as-welded specimen exhibited the local necking and rupture in the BMZ. In addition, the Type-II filler material with a greater value of equivalent carbon content was rarely influenced by the tempering, maintaining its hardness in the as-quenched status, while the Type-I filler material showed a gradual decrease in hardness with the tempering time. The results demonstrate that the Type-II weld joint outperformed the Type-I weld joint in terms of the structural integrity of welded parts.

Thyroid cancer among female workers in Korea, 2007-2015.

BACKGROUND: Prevalence of thyroid cancer has been increasing rapidly worldwide, especially among women. There has been a debate as to whether such an increase represents consequences of over-diagnosis or a true increase. To find the occupational risk of Korean female workers in different industry sectors, we analyzed the data of Korean female workers. METHODS: National Female Worker Cohort data that contain information on total female workers were used for our analysis of prevalence of thyroid cancer (C73 according to KCD-5, 6 code) derived from National Health Insurance data. By combining industrial codes from National Health Insurance Service and those from Korea Workers' Compensation and Welfare Service, the classification of industrial codes became to consist of the total of thirty three representing both non-office (NO) and office (O) categories. Both an internal comparison among female workers within the cohort and an external comparison to compare female workers with Korean general female population were carried out. RESULTS: Among 149,258 female workers, 2,641 cases of thyroid cancers were identified. Differences in prevalence of thyroid cancer between female workers (40.5%) and general Korean female population (32.6%) were observed; however, the differences in prevalence of thyroid cancer between NO workers and O workers were not apparent. An analysis involving workers in Financial and insurance activities sector revealed that, standardized prevalence rate (SPR) of both NO (2.96, 95% CI = 2.01-4.20) and O workers (3.68, 95% CI = 3.10-4.33) increased significantly and that an AOR (adjusted odds ratio) increased marginally (1.38, 95% CI = 0.97-1.96). Further, when stratified in respect to the duration of employment, an AOR of female workers having been employed for more than 8 years showed a significant increase (1.63, 95% CI = 1.07-2.49). CONCLUSIONS: Female workers had a higher risk of thyroid cancer than general female population but the difference between NO workers and O workers was not found to be significant in most industrial sectors. Further studies using data with information regarding specific occupational exposures are needed.

Health Effects from Occupational Radiation Exposure among Fluoroscopy-Guided Interventional Medical Workers: A Systematic Review.

A systematic review was conducted to provide an overview of the health effects of occupational radiation exposure from interventional fluoroscopy procedures on medical radiation workers. Among the 34 studies that met the inclusion criteria, most studies were cross-sectional (76%) and published after 2011 (65%) in a handful of countries. Although diverse outcomes were reported, most studies focused on cataracts. Radiation health effects were rarely assessed by risk per unit dose. Interventional radiation medical workers represent a small subset of the population studied worldwide. Further epidemiologic studies should be conducted to evaluate health outcomes among interventional radiation medical workers.

What causes the experience of discrimination in non-regular workers?

BACKGROUND: Discrimination based on type of employment against non-regular workers is still a social issue. However, there are few studies on job factors that affect the discrimination experience in each type of employment or the association between discrimination and health impact indicators. This study examined occupational health characteristics according to discrimination experience and relating factors that affect discrimination experience. METHODS: This study used the 4th Korean Working Conditions Survey (2014) provided by the Korea Occupational Safety and Health Agency. Among the 50,000 workers, 7731 non-regular wage workers were selected as study population. To examine differences in discrimination experience, we used a t-test on occupational risk factors, occupational stress, occupational characteristics, health impact indicators. To identify the factors that affected discrimination experience, we performed binomial logistic regression analysis. RESULTS: The discrimination experience rate was significantly higher in male, aged less than 40 years old, above high school graduate than middle school graduate, higher wage level, shorter employment period and larger company's scale. As factors related to discrimination experience, they experienced discrimination more as occupational stress was higher and when they were temporary or daily workers rather than permanent workers, work patterns were not consistent, and the support of boss was low. It showed that physical, musculoskeletal, and mental occupational risk scores and subjective job instability were higher and work environment satisfaction was lower in discrimination experienced group. CONCLUSIONS: The present study showed that the demographic and occupational factors were complexly related to discrimination experience in non-regular workers. The experience of discrimination had increased when occupational stress was higher, they were temporary or daily workers rather than permanent workers, work patterns were not consistent, and their boss' support was low. Improving various relating factors, (e.g. occupational stresses, employment status and occupational characteristics), this would ultimately expect to improve non-regular workers' discrimination.

Relating factors to wearing personal radiation protectors among healthcare professionals.

BACKGROUND: With increasing use of medical radiologic procedures, wearing proper protector should be emphasized to reduce occupational radiation exposures. This research describes the rates of lead apron wearing for radiation protection and assessed occupational factors related to wearing rates for various types of healthcare professionals. METHODS: We conducted a self-administered questionnaire survey through a website, on-site visits, fax, and mail. Of the 13,489 participants, 8858 workers who could not completely separate themselves from radiological procedure areas. Their general characteristics (sex and age), work history (job title, duration of employment, and hospital type), and practices (frequency of radiation procedures, ability to completely separate from radiation, and frequency of wearing protective lead aprons) were examined. RESULTS: The mean rate of lead apron wearing during radiologic procedures was 48.0 %. The rate was different according to sex (male: 52.9 %, female: 39.6 %), hospital type (general hospital: 63.0 %, hospital: 51.3 %, clinic: 35.6 %, dental hospital/clinic: 13.3 %, public health center: 22.8 %), and job title (radiologic technologist: 50.3 %, doctor: 70.3 %, dentist/dental hygienist: 15.0 %, nurse/nursing assistant: 64.5 %) (p < 0.001). By logistic regression analysis stratified by job title, use of lead aprons by radiologic technologists and nurses/nursing assistants was associated with hospital type and exposure frequency score. For doctors, apron wearing was associated with employment duration. For dentists/dental hygienists, apron wearing was associated with the exposure frequency score. CONCLUSIONS: To improve working environments for healthcare professionals exposed to radiation, it is necessary to consider related factors, such as job title, duration of employment, and hospital type, when utilizing a planning and management system to prevent radiation-related health problems.

Adaptive characterization of recrystallization kinetics in IF steel by electron backscatter diffraction.

In this study, a rigorous methodology for quantifying recrystallization kinetics by electron backscatter diffraction is proposed in order to reduce errors associated with the operator's skill. An adaptive criterion to determine adjustable grain orientation spread depending on the recrystallization stage is proposed to better identify the recrystallized grains in the partially recrystallized microstructure. The proposed method was applied in characterizing the microstructure evolution during annealing of interstitial-free steel cold rolled to low and high true strain levels of 0.7 and 1.6, respectively. The recrystallization kinetics determined by the proposed method was found to be consistent with the standard method of Vickers microhardness. The application of the proposed method to the overall recrystallization stages showed that it can be used for the rigorous characterization of progressive microstructure evolution, especially for the severely deformed material.

Contrast spreading patterns in retrodiscal transforaminal epidural steroid injection.

OBJECTIVE: To observe the contrast spreading patterns in the retrodiscal (RD) approach for transforaminal epidural steroid injections and their effect on pain reduction. METHOD: Patients with L5 radiculopathy who were scheduled to receive lumbar TF-EPB were consecutively included. We randomly divided them into the L4-5 RD and L5-S1 RD groups and administered 1 cc of contrast dye into epidural space. We observed the shape and the location of contrast dye on the anterior-posterior and lateral views. We injected 1 cc of 0.5% lidocaine mixed with 20 mg of triamcinolone, and checked the pain intensity before and two weeks after the procedure by using visual analogue scale (VAS). RESULTS: In the L4-5 RD group (n=30), contrast spread over the L4 nerve root in 27 cases and the L4 and L5 nerve roots in 3 cases. In the L5-S1 RD group (n=33), contrast spread over the L5 nerve root in 20 cases, the S1 nerve root in 3 cases, and the L5 and the S1 nerve roots in 10 cases. The contrast spreading patterns could be divided into 4 patterns: the proximal root in 40 cases, the distal root in 19 cases, the anterior epidural space in 3 cases and an undefined pattern in 1 case. CONCLUSION: In RD lumbar TF-EPB, the contrast dye mostly went into the cephalic root and about 60% spread over the proximal nerve root. There was less pain reduction when the contrast dye spread over the distal nerve root.