• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1153-1160
• /
• 2021

This paper proposes a routing algorithm that determines the optimal path using deep reinforcement learning in software-defined networks. The deep reinforcement learning model for learning is based on DQN, the inputs are the current network state, source, and destination nodes, and the output returns a list of routes from source to destination. The routing task is defined as a discrete control problem, and the quality of service parameters for routing consider delay, bandwidth, and loss rate. The routing agent classifies the appropriate service class according to the user's quality of service profile, and converts the service class that can be provided for each link from the current network state collected from the SDN. Based on this converted information, it learns to select a route that satisfies the required service level from the source to the destination. The simulation results indicated that if the proposed algorithm proceeds with a certain episode, the correct path is selected and the learning is successfully performed.

• Korean Journal of Schizophrenia Research
• /
• v.24 no.2
• /
• pp.60-67
• /
• 2021

Objectives: The optimal duration of maintenance treatment for patients with first-episode schizophrenia (FES) remains unclear. We examined the first antipsychotic treatment duration and its association with re-initiation of treatment using a nationwide claim database. Methods: Data from the Health Insurance Review and Assessment Service database in South Korea for 2007-2016 were used. Linear regression analysis and Cox proportional hazard models were used to evaluate the associations between the duration of the first antipsychotic treatment, time to re-initiation of treatment, and occurrence of treatment re-initiation. Results: Of 30,143 patients with FES, 80.4% (n=24,231) received <2 years of the first antipsychotic treatment. In patients who discontinued treatment (n=23,030), the rate of treatment re-initiation was 74.2% (n=17,086). As the duration of the first antipsychotic treatment increased, the time to re-initiation of treatment decreased (β=-0.146, p<0.001); however, the rate of treatment reinitiation was relatively constant (hazard ratio=1.001, p<0.001). Conclusion: Long-term antipsychotic treatment was not significantly associated with the rate of treatment re-initiation but showed a negative association with the time to re-initiation of treatment. Further research is needed to better understand the optimal treatment duration for FES.

• International Journal of Computer Science & Network Security
• /
• v.22 no.6
• /
• pp.131-138
• /
• 2022

Novel Coronavirus (COVID-19) is viewed as one of the main general wellbeing theaters on the worldwide level all over the planet. Because of the abrupt idea of the flare-up and the irresistible force of the infection, it causes individuals tension, melancholy, and other pressure responses. The avoidance and control of the novel Covid pneumonia have moved into an imperative stage. It is fundamental to early foresee and figure of infection episode during this troublesome opportunity to control of its grimness and mortality. The entire world is investing unimaginable amounts of energy to fight against the spread of this lethal infection. In this paper, we utilized machine learning and deep learning techniques for analyzing what is going on utilizing countries shared information and for detecting the climate factors that effect on spreading Covid-19, such as humidity, sunny hours, temperature and wind speed for understanding its regular dramatic way of behaving alongside the forecast of future reachability of the COVID-2019 around the world. We utilized data collected and produced by Kaggle and the Johns Hopkins Center for Systems Science. The dataset has 25 attributes and 9566 objects. Our Experiment consists of two phases. In phase one, we preprocessed dataset for DL model and features were decreased to four features humidity, sunny hours, temperature and wind speed by utilized the Pearson Correlation Coefficient technique (correlation attributes feature selection). In phase two, we utilized the traditional famous six machine learning techniques for numerical datasets, and Dense Net deep learning model to predict and detect the climatic factor that aide to disease outbreak. We validated the model by using confusion matrix (CM) and measured the performance by four different metrics: accuracy, f-measure, recall, and precision.

• Physical Therapy Korea
• /
• v.29 no.1
• /
• pp.54-63
• /
• 2022

Background: Low back and neck pain are two of the most common medical problems in the adult population. It is estimated that between 70% and 80% adults experience an episode of low back pain at least once during their lifetime. Objects: This study aimed to compare the effects of various stretching exercises and muscle energy techniques used for treatment of iliopsoas, quadratus lumborum, and hamstrings of patients with low back pain. Methods: A total of 52 subjects were randomly assigned to the control group (n = 17), the muscle energy technology group (METG, n = 19), and the stretching exercise group (SEG, n = 16). Interventions were performed twice a week for 4 weeks. Outcomes were measured before and after 4 weeks of treatment in the three groups, using the Korean version of the Oswestry Disability Index, the Visual Analogue Scale (VAS), the Fear Avoidance Belief Questionnaire (FABQ), the Finger to Floor test (FTF), the Modified-Modified Schober Test (MMS), and the Trunk Flexion Range of Motion test. Results: This study showed significant reduction in the pre-test and post-test KODI, VAS, and FABQ scores in all the three groups (p < 0.05). The KODI, VAS, FABQ and FTF assessed in this study showed interactions between the groups and the measurement time point (p < 0.01). After intervention, the KODI, VAS, FABQ and FTF in the SEG and METG were significantly higher (p < 0.05), and there was no difference between the METG and SEG. MMS and HFROM test showed no difference between the three groups (p > 0.05). Conclusion: KODI, VAS, FABQ and FTF showed significant improvement after basic physical therapy, muscle energy technique, and stretching exercise intervention. And there was no significant difference in the intervention effect between the muscle energy group and the stretching exercise group.

• Journal of Oriental Neuropsychiatry
• /
• v.32 no.4
• /
• pp.329-335
• /
• 2021

Objectives: The purpose of this study was to determine how analyzing movies, including mental symptoms, according to a certain method, might affect the understanding of psychiatric disorders. Methods: Forty-four oriental medicine students who had completed Korean medicine neuropsychiatric science were required to submit reports on episode analysis, psychiatric personal history investigation, diagnostic criteria, and the connection between Korean medicine and psychiatric diseases after watching a movie, including psychiatric symptoms. After submitting the report, a questionnaire related to understanding before and after watching the movie was asked to be filled out. Demographic survey, frequency analysis, and response sample t-test were performed based on 42 questionnaires. Results: Results of analyzing the questionnaire were as follows. 1. The average number of movies watched was three. 2. Psychiatric disorders and psychiatric symptoms, diagnostic criteria, psychiatric personal investigation, and understanding of the connection between Korean medicine and psychiatric diseases all increased statistically significantly. 3. A separate process might be needed to improve the understanding of psychiatric personal strength investigation and oriental medicine connection. Conclusions: Movie analysis, including individual mental symptoms, could improve students' understanding of psychiatric disorders in psychiatric symptoms, diagnostic criteria, and psychiatric personal investigation, but some students might need feedback.

• Journal of Environmental Science International
• /
• v.32 no.8
• /
• pp.521-531
• /
• 2023

This study analyzed the impact of recirculation on high-concentration PM_2.5 in the coastal area. Through the analysis of observational data, it was observed that the development of sea breeze led to an increase in PM_2.5 and SO₄^2- concentrations. Hysplit backward trajectory analysis confirmed the occurrence of air mass recirculation. Results from WRF and CMAQ numerical simulations indicated that pollutants transported from land to sea during the night were re-transported to the land by daytime sea breeze, leading to high-concentration PM_2.5 in Busan. To quantitatively investigate the recirculation a recirculation factor (RF) was calculated, showing an increase in RF values during high-concentration PM_2.5 episodes. However, the RF values varied slightly depending on the time resolution of meteorological data used for the calculations. This variation was attributed to the terrain characteristics at observation sites. Additionally, during long-range transported days leading to nationwide high-concentration PM_2.5 events, synoptic-scale circulation dominated, resulting in weaker correlation between PM_2.5 concentration and RF values. This study enhances the understanding of the influence of recirculation on air pollution. However, it is important to consider the impact of temporal resolution and terrain characteristics when using RF for evaluating recirculation during episodes of air pollution.

• The korean journal of orthodontics
• /
• v.53 no.5
• /
• pp.298-306
• /
• 2023

Objective: Pain following the insertion of separators and archwires varies with age, sex, race, ethnicity, threshold, and health status. This study aimed to evaluate the characteristics of pain in orthodontic patients after the insertion of elastomeric separators, its effects on daily life, and its association with age and sex in a population not previously studied in this regard. Methods: A cross-sectional study of 130 patients undergoing orthodontic treatment included collecting data on demographics, pain experienced following the placement of separators, time of onset, duration, characteristics, change in dietary pattern or chewing side, intake of analgesics, and severity of pain on the Wong Baker's scale. The results are reported as counts and percentages. Associations between sex and age were evaluated using Pearson's chi-square test. Results: Among the 130 patients, 56.2% were 9-20 years old, 63.8% experienced pain following the insertion of separators, 22.9% had their first episode of pain at 4 hours, 56.6% experienced intermittent pain, and 37.3% experienced discomfort; 18.1% males and 81.9% females experienced pain following the insertion of separators. Pearson's chi-square test showed a significant association between pain and sex (P = 0.04). Most patients (37.3%) reported "hurts little more" for pain intensity on Wong Baker's scale and 21.7% reported all four quadrants as sites of pain. Conclusions: The pain experienced after separator insertion was associated with sex and age. Females experienced more pain than males and patients between the age range of 21 and 36 years suffered more pain during mastication than between 9 and 20 years old.

• Journal of Trauma and Injury
• /
• v.36 no.1
• /
• pp.39-48
• /
• 2023

Purpose: Sports offer several health benefits but are not free of injury risk. Activity dynamics vary across sports, impacting the injury profile and thereby influencing healthcare resource utilization and health outcomes. The purpose of this study was to investigate sports-related major trauma cases and compare differences across sports and activity groups. Methods: A retrospective case notes review of sports-related major traumas over a 5-year period was conducted. Demographic, hospital episode-related, and health outcome-related data were analyzed, and differences were compared across sports and activity groups. The Glasgow Outcome Scale (GOS) at discharge was used as the primary outcome measure and the length of hospital stay as the secondary outcome measure. Results: In total, 76% of cases had good recovery at discharge (GOS, 5), 19% had moderate disability (GOS, 4), and 5% had severe disability (GOS, 3). The mean length of hospital stay was 11.2 days (range, 1-121 days). The most severely injured body region was the limbs (29.1%) and vertebral/spinal injuries were most common (33%) in terms of location. A significant difference (P<0.05) existed in GOS across sports groups, with motor sports having the lowest GOS. However, no significant differences (P>0.05) were found in other health-outcome variables or injury patterns across sports or activity groups, although more competitive sports cases (67%) required admission than recreational sports cases (33%). Conclusions: Spinal injuries are the most frequent sports injuries, bear the worst health outcomes, and warrant better preventive measures. Head injuries previously dominated the worst outcomes; this change is likely due to better preventive and management modalities. Competitive sports had a higher injury frequency than recreational sports, but no difference in health outcomes or injury patterns.

• Atmosphere
• /
• v.33 no.1
• /
• pp.21-32
• /
• 2023

Characteristics of downslope windstorm (DW) has been examined mainly based on 1-min average wind and the other meteorological conditions in the Yeongdong region for 2000~2020. First, a classification procedure for the downslope windstorm is proposed using surface wind speed (greater than 99 percentile), 1-hour longevity of strong wind (SW), westerly wind direction, low humidity (less than 20 percentile), and leeside warming. The number of DW days satisfying the proposed criteria is 221 (2.9% of total days and 47.5% of SW days) while the number of SW days is 465 (6.1% of total days) for 2000~2020. The occurrences of both SW and DW shows distinctive annual variation with its peak in April. In addition, mean wind speed of DW days is 8.2 m s^-1 with its duration of 2 hr 30 min and relative humidity of 28% at Gangneung. An episode (7 May 2021) was selected by applying the proposed criteria to SW days of 2021. The sounding shows that the layer of wind speed greater than 25 m s^-1 was lowered down to 925 hPa at Gangneung (leeside) relative to 850 hPa at Hoengseong (Wonju), in the afternoon along with significant warming and drying. Froude numbers of Wonju and Gangneung for the DW events were increased 4 and 5 times greater than those of normal days, respectively. This kind of DW long-term statistics in the leeside of the mountains is thought to build a foundation of further understanding DW mechanism.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.75-81
• /
• 2022

Autonomous vehicles, which can dramatically solve the lack of parking spaces, are making great progress through deep reinforcement learning. Activation functions are used for deep reinforcement learning, and various activation functions have been proposed, but their performance deviations were large depending on the application environment. Therefore, finding the optimal activation function depending on the environment is important for effective learning. This paper analyzes 12 functions mainly used in reinforcement learning to compare and evaluate which activation function is most effective when autonomous vehicles use deep reinforcement learning to learn parking. To this end, a performance evaluation environment was established, and the average reward of each activation function was compared with the success rate, episode length, and vehicle speed. As a result, the highest reward was the case of using GELU, and the ELU was the lowest. The reward difference between the two activation functions was 35.2%.