• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.6
• /
• pp.33-45
• /
• 2020

This paper aims to look at the perspective that the latest cutting-edge technologies are predicting individual diseases in the actual medical environment in a situation where various types of wearable devices are rapidly increasing and used in the healthcare domain. Through the process of collecting, processing, and transmitting data by merging clinical data, genetic data, and life log data through a user-participating wearable device, it presents the process of connecting the learning model and the feedback model in the environment of the Deep Neural Network. In the case of the actual field that has undergone clinical trial procedures of medical IT occurring in such a high-tech medical field, the effect of a specific gene caused by metabolic syndrome on the disease is measured, and clinical information and life log data are merged to process different heterogeneous data. That is, it proves the objective suitability and certainty of the deep neural network of heterogeneous data, and through this, the performance evaluation according to the noise in the actual deep learning environment is performed. In the case of the automatic encoder, we proved that the accuracy and predicted value varying per 1,000 EPOCH are linearly changed several times with the increasing value of the variable.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1579-1590
• /
• 2020

In this paper, spatially-adaptive denormalization (SPADE) based U-Net is proposed to detect changes by using high-resolution satellite images. The proposed network is to preserve spatial information using SPADE. Change detection methods using high-resolution satellite images can be used to resolve various urban problems such as city planning and forecasting. For using pixel-based change detection, which is a conventional method such as Iteratively Reweighted-Multivariate Alteration Detection (IR-MAD), unchanged areas will be detected as changing areas because changes in pixels are sensitive to the state of the environment such as seasonal changes between images. Therefore, in this paper, to precisely detect the changes of the objects that consist of the city in time-series satellite images, the semantic spatial objects that consist of the city are defined, extracted through deep learning based image segmentation, and then analyzed the changes between areas to carry out change detection. The semantic objects for analyzing changes were defined as six classes: building, road, farmland, vinyl house, forest area, and waterside area. Each network model learned with KOMPSAT-3A satellite images performs a change detection for the time-series KOMPSAT-3 satellite images. For objective assessments for change detection, we use F1-score, kappa. We found that the proposed method gives a better performance compared to U-Net and UNet++ by achieving an average F1-score of 0.77, kappa of 77.29.

• Therapeutic Science for Rehabilitation
• /
• v.10 no.3
• /
• pp.23-41
• /
• 2021

Objective : The purpose of this study was to analyze the effects of simultaneous dual-task training to assess executive function in older adults. Methods : We searched the PubMed, EMBASE, Cochrane, Web of Science, and RISS databases of publicated studies in the past decade. Seven studies were selected based on the inclusion and exclusion criteria. Qualitative assessment and meta-analysis were performed for the seven studies. Results : A randomized controlled trial design was used in the selected studies, and PEDro Scores above seven were obtained. The Trial Making Test (TMT) evaluated the effects of dual-task training on executive function in four studies. The Color Trail Test (CTT) was used in two studies, and Stroop test was used in three studies. The effect size for total executive function was 0.38, which was small. The effect sizes for TMT and CTT were 0.37. Stroop Test was 0.34, demonstrating that their effect sizes were also small. Only significant effects in total executive function, TMT, and CTT showed significant effects (all p<0.05). Conclusion : This study confirmed that dual-task training was effective in improving executive function in older adults. To improve the effectiveness of dual-task training, the difficulty of the dual-task training should be considered. It is also necessary to implement assessments that can evaluate performance under dual-task conditions as well as conventional test tools for executive function. In the future, dual-task training could be used as an appropriate intervention for executive function in older adults to delay the onset of dementia.

• The korean journal of orthodontics
• /
• v.52 no.1
• /
• pp.3-19
• /
• 2022

Objective: The purpose of this study was to investigate the accuracy of one-step automated orthodontic diagnosis of skeletodental discrepancies using a convolutional neural network (CNN) and lateral cephalogram images with different qualities from nationwide multi-hospitals. Methods: Among 2,174 lateral cephalograms, 1,993 cephalograms from two hospitals were used for training and internal test sets and 181 cephalograms from eight other hospitals were used for an external test set. They were divided into three classification groups according to anteroposterior skeletal discrepancies (Class I, II, and III), vertical skeletal discrepancies (normodivergent, hypodivergent, and hyperdivergent patterns), and vertical dental discrepancies (normal overbite, deep bite, and open bite) as a gold standard. Pre-trained DenseNet-169 was used as a CNN classifier model. Diagnostic performance was evaluated by receiver operating characteristic (ROC) analysis, t-stochastic neighbor embedding (t-SNE), and gradient-weighted class activation mapping (Grad-CAM). Results: In the ROC analysis, the mean area under the curve and the mean accuracy of all classifications were high with both internal and external test sets (all, > 0.89 and > 0.80). In the t-SNE analysis, our model succeeded in creating good separation between three classification groups. Grad-CAM figures showed differences in the location and size of the focus areas between three classification groups in each diagnosis. Conclusions: Since the accuracy of our model was validated with both internal and external test sets, it shows the possible usefulness of a one-step automated orthodontic diagnosis tool using a CNN model. However, it still needs technical improvement in terms of classifying vertical dental discrepancies.

• Journal of Korean Neurosurgical Society
• /
• v.65 no.2
• /
• pp.287-296
• /
• 2022

Objective : Although radiotherapy (RT) is recommended for multiple myeloma (MM) involving spine, the treatment of choice between reconstructive surgery with RT and RT alone for pathologic vertebral fractures (PVFs) associated with structural instability or neurologic compromises remains controversial. The purpose of this study was to evaluate the clinical efficacies of reconstructive surgery with adjuvant RT for treatment of MM with PVFs by comparing with matched cohorts treated with RT alone. Methods : Twenty-eight patients underwent reconstructive surgery followed by RT between 2008 and 2015 in a single institution, for management of PVFs associated with structural instability of the spine and/or neurologic compromises (group I). Twenty-eight patients were treated with RT alone (group II) after propensity score matching in a 1-to-1 format based on instability of the spine, as well as age and performance. Clinical outcomes including the overall survival rates, duration of independent ambulation, neurological status, and numeric rating scale (NRS) for back pain were compared. Results : Clinical and radiological features before treatment were similar in both groups. The median survival period was similar between the two groups. However, the mean duration of independent ambulation was significantly longer in group I (88.8 months; 95% confidence interval [CI], 66.0-111.5) than in group II (39.4 months; 95% CI, 25.2-53.6) (log rank test; p=0.022). Deterioration of Frankel grade (21.4% vs. 60.7%, p=0.024) and NRS for back pain (2.7±2.2 vs. 5.0±2.7, p=0.000) at the last follow-up were higher in the group II. Treatment-related complications were similar in both groups. Conclusion : In patients with unstable PVFs due to MM, reconstructive surgery may yield superior clinical outcomes compared with RT alone in maintaining independent ambulation and neurological status, as well as pain control despite similar median survival and complications.

• Health Policy and Management
• /
• v.32 no.1
• /
• pp.94-106
• /
• 2022

Background: The efforts to build more "people centered," "patient centered" health system has been emerging all over the world. Aligning with it, the Korean government is conducing the survey called "Medical Service Experience Survey (MSES)." There are critics, however, that MSES is not scrutinizing the medical experiences of patients in various healthcare settings. For this reason, this study aims to perform an empirical analysis of the differences in answers of patients responding to various healthcare settings. Methods: There are two steps in this study. First, explanatory analysis is conducted to compare the tendency of statistical concentration on questionnaires by divided healthcare settings. Second, confirmative analysis is carried out to evaluate the construct validity, reliability, and discriminant validity of the questionnaire in each healthcare setting. The raw data of MSES, which was conducted in 2020 by the Ministry of Health and Welfare in Korea and the Korean Institute for Health and Social Affairs is used. Results: As a result of exploratory factor analysis for all outpatients, the items were classified into four factors statistically: "doctor experience," "nurse experience," "outpatient service experience," and "patient satisfaction." It was confirmed that the reliability of all factors extracted was secured. However, for patients who visited hospitals, questionnaires related to personal privacy, such as "experiences on medical staffs considering physical exposure" or "experiences related to personal information exposure," were answered in conjunction with items of "nurse experience." Besides, patients responded that administrative elements of medical services, such as "experiences of comfort in medical institutions" and "experiences of satisfactory administrative services," were related to the items of "nurse experience." The answers of patients who visited traditional medical hospitals and clinics about "doctor experience" and "nurse experience" were not discerned statistically, and the answers to "doctor experience," "nurse experience," and "medical institution experience" were entangled with the responses of patients who visited dental hospitals and clinics. On the other hand, as a result of the confirmatory factor analysis, it was found that the inquiries of MSES generally had intensive validity. Conclusion: The collection of objective and scientific data is the prominent component to enlighten the patient-centered healthcare system alongside with change of the worldwide paradigm of measuring the healthcare system performance as follows the transition of perspective of health care from provider-centered to patient-centered. This study empirically shows that the patient experience can vary as the healthcare settings. Furthermore, to make an advance in measuring the experience of patients with medical services, this article proposes the deliberate consideration of the different kinds of healthcare settings and articulate design of the survey.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1643-1652
• /
• 2022

Deposited Marine Debris(DMD) can negatively affect marine ecosystems, fishery resources, and maritime safety and is mainly detected by sonar sensors, lifting frames, and divers. Considering the limitation of cost and time, recent efforts are being made by integrating underwater images and artificial intelligence (AI). We conducted a comparative study of You Only Look Once Version 5 (YOLOv5) and You Only Look Once Version 7 (YOLOv7) models to detect DMD from underwater images for more accurate and efficient management of DMD. For the detection of the DMD objects such as glass, metal, fish traps, tires, wood, and plastic, the two models showed a performance of over 0.85 in terms of Mean Average Precision (mAP@0.5). A more objective evaluation and an improvement of the models are expected with the construction of an extensive image database.

• Animal Bioscience
• /
• v.35 no.11
• /
• pp.1733-1743
• /
• 2022

Objective: This study was aimed to explore the efficacy of combination of endo-xylanase (Xyn) and xylan-debranching enzymes (arabinofuranosidase, Afd and feruloyl esterase, FE) in improving utilization of bran in piglet diet. Methods: In vitro experiments were firstly conducted to examine the enzymological properties of Xyn, Afd, and FE, concurrent with their effect on degradation of arabinoxylan (Abx) in bran. In vivo experiment was then implemented by allocating two hundred and seventy 35-d-old postweaning piglets into 3 groups (6 replicates/group), which received bran-containing diet supplemented with Xyn (1,600 U/kg) or its combination with Afd (0.8 U/kg) and FE (4 U/kg) or without enzyme. Results: Both Xyn, Afd, and FE are relatively stable against the changes in temperature and pH value. Combining Xyn with Afd and FE had a superiority (p<0.05) over Xyn alone and its combination with Afd or FE in promoting (p<0.05) degradation of Abx in different brans. Combined treatment with Xyn, Afd, and FE was more beneficial than Xyn alone to induce increasing trends (p<0.10) of average daily gain, final body weight and feed efficiency of piglets fed bran-containing diet. Moreover, combination of Xyn, Afd, and FE showed advantages (p<0.05) over Xyn alone in causing reductions (p<0.05) in diarrhea rate and cecal pH value, concurrent with increases (p<0.05) in cecal and colonic acetic acid and total volatile fatty acid concentrations, as well as cecal butyric acid concentration of piglets fed bran-containing diet. Conclusion: Combining Xyn with Afd and FE was more beneficial than Xyn alone in promoting degradation of Abx in bran, along with growth performance and intestinal volatile fatty acid profile of piglets received bran-containing diet. Thereby, combination of Xyn, Afd, and FE had a superior efficacy relative to Xyn alone in improving application of cereal bran in piglet diet.

• Animal Bioscience
• /
• v.35 no.11
• /
• pp.1752-1759
• /
• 2022

Objective: Effects of direct-fed Enterococcus faecium plus bacteriophages (EF-BP) were investigated as potential substitutes for pharmacological ZnO for weanling pigs. Methods: Dietary treatments were supplementations to a basal diet with none (NC), 3,000-ppm ZnO (PC), 1×10¹⁰ colony-forming units of E. faecium plus 1×10⁸ plaque-forming units (PFU) of anti-Salmonella typhimurium bacteriophages (ST) or 1×10⁶ PFU of each of anti-enterotoxigenic Escherichia coli K88 (F4)-, K99 (F5)-, and F18-type bacteriophages (EC) per kg diet. In Exp 1, twenty-eight 21-day-old crossbred weanling pigs were individually fed one of the experimental diets for 14 days and euthanized for histological examination on intestinal mucosal morphology. In Exp 2, 128 crossbred weanling pigs aged 24 days were group-fed the same experimental diets in 16 pens of 8 piglets on a farm with a high incidence of post-weaning diarrhea. Results: None of the diarrheal score or fecal consistency score (FCS), average daily gain (ADG), gain: feed ratio, structural variables of the intestinal villus, and goblet cell density, differed between the EF-BP (ST+EC) and NC groups, between EF-BP and PC, or between ST and EC, with the exception of greater gain: feed for EF-BP than for PC (p<0.05) during days 7 to 14 (Exp 1). In Exp 2, ADG was less for EF-BP vs PC during days 0 to 7 and greater for EF-BP vs NC during days 7 to 14. FCS peaked on day 7 and declined by day 14. Moreover, FCS was less for EF-BP vs NC, did not differ between EF-BP and PC, and tended to be greater for ST vs EC (p = 0.099). Collectively, EF-BP was comparable to or slightly less effective than PC in alleviating diarrhea and growth check of the weanling pigs, with ST almost as effective as PC, when they were group-fed. Conclusion: The E. faecium-bacteriophage recipe, especially E. faecium-anti-S. typhimurium, is promising as a potential substitute for pharmacological ZnO.

• Animal Bioscience
• /
• v.35 no.11
• /
• pp.1760-1770
• /
• 2022

Objective: It is crucial to accurately determine the net energy (NE) values of feed ingredients because the NE system is expected to be applied to the formulation of broilers feed. The NE values of 5 wheat and 5 wheat brans were determined in 12-to 14-day old Arbor Acres (AA) broilers with substitution method and indirect calorimetry method. Methods: A total of 12 diets, including 2 reference diets (REF) and 10 test diets (5 wheat diets and 5 wheat bran diets) containing 30% of test ingredients, were randomly fed to 864 male AA birds with 6 replicates of 12 birds per treatment. These birds were used to determine metabolizable energy (ME) (8 birds per replicate) in the chicken house and NE (4 birds per replicate) in the chamber respectively at the same time. After a 4-d dietary and environment adaptation period, growth performance, energy values, energy balance and energy utilization were measured during the following 3 d. Multiple linear regression analyses were further performed to generate prediction equations for NE values based on the chemical components and ME values. The NE prediction equation were also validated on another wheat diet and another wheat bran diet with high correlation (r = 0.98, r = 0.75). Results: The NE values of 5 wheat and 5 wheat bran samples are 9.34, 10.02, 10.27, 11.33, and 10.49 MJ/kg, and 5.37, 5.17, 4.87, 5.06, and 4.88 MJ/kg DM, respectively. The equation with the best fit were NE = 1.968AME-0.411×ADF-14.227 (for wheat) and NE = -0.382×CF-0.362×CP-0.244×ADF+20.870 (for wheat bran). Conclusion: The mean NE values of wheat and wheat bran are 10.29 and 5.07 MJ/kg DM in AA broilers. The NE values of ingredients could be predicted by their chemical composition and energy value with good fitness.