Search

Precise remote evaluation of both suicide risk and psychiatric disorders is critical for suicide prevention as well as for psychiatric well-being. Using questionnaires is an alternative to labor-intensive diagnostic interviews in a large general population, but previous models for predicting suicide attempts suffered from low sensitivity. We developed and validated a deep graph neural network model that increased the prediction sensitivity of suicide risk in young adults (n = 17,482 for training; n = 14,238 for testing) using multi-dimensional questionnaires and suicidal ideation within 2 weeks as the prediction target. The best model achieved a sensitivity of 76.3%, specificity of 83.4%, and an area under curve of 0.878 (95% confidence interval, 0.855-0.899). We demonstrated that multi-dimensional deep features covering depression, anxiety, resilience, self-esteem, and clinico-demographic information contribute to the prediction of suicidal ideation. Our model might be useful for the remote evaluation of suicide risk in the general population of young adults for specific situations such as the COVID-19 pandemic., (© 2021. The Author(s).)

This study aimed to examine the health effect of secondhand smoke (SHS) exposure at home, school, and/or public places on allergic multimorbidity using nationwide data among school-attending adolescents in Korea. Allergic multimorbidity was defined as two or more coexisting allergic diagnoses of asthma, allergic rhinitis, and/or atopic dermatitis during the past 12 months. A multinomial logistic regression analysis was performed to evaluate the association of SHS exposure and allergic multimorbidity. Of the study participants, 24.3% were diagnosed as having any allergic disease currently and 66.3% reported SHS exposure. Any SHS exposure that includes public places conferred increased odds of atopic dermatitis in non-current smokers (adjusted odds ratio 1.21-1.46; 95% confidence interval [CI] 1.10-1.66). Moreover, when controlling for current smoking additionally, SHS exposure at the three sites was 1.37 and 1.96 times more likely to be associated with allergic single and multiple morbidities, respectively (95% CI 1.26-1.49 and 1.65-2.31, respectively). In conclusion, this study found positive associations of SHS exposure with single or multiple allergic morbidity compared to no exposure at all. Further studies with longitudinal designs and objective measurement of SHS exposure and allergic diagnosis are warranted.

To investigate associations of the duration of voriconazole treatment and radiological response with relapse of invasive pulmonary aspergillosis (IPA) in immunocompromised patients, we explored the risk factors for IPA relapse after successful initial treatment. All patients with proven or probable IPA who had finished voriconazole treatment between 2005 and 2019 in a tertiary-care hospital were reviewed. IPA relapse was defined as re-diagnosis of proven or probable IPA at the same site within 1 year after treatment termination. Short course of voriconazole treatment was defined as a treatment less than 9 weeks, which is a median of the recommended minimum duration of therapy from the Infectious Disease Society of America. The radiological response was defined as a reduction in IPA burden by more than 50% on chest computed tomography. Of 87 patients who had completed voriconazole treatment, 14 (16.1%) experienced IPA relapse. Multivariable Cox regression identified that short voriconazole treatment duration (adjusted hazard ratio [aHR], 3.7; 95% confidence interval [CI], 1.1-12.3; P = 0.033) and radiological non-response (aHR, 4.6; 95% CI, 1.2-17.5; P = 0.026) were independently associated with relapse of IPA after adjusting for several clinical risk factors. Longer duration of therapy should be considered for those at higher risk of relapse.

The purpose of this study was to evaluate and compare the diagnostic performances of the deep convolutional neural network (CNN) and expert radiologists for differentiating thyroid nodules on ultrasonography (US), and to validate the results in multicenter data sets. This multicenter retrospective study collected 15,375 US images of thyroid nodules for algorithm development (n = 13,560, Severance Hospital, SH training set), the internal test (n = 634, SH test set), and the external test (n = 781, Samsung Medical Center, SMC set; n = 200, CHA Bundang Medical Center, CBMC set; n = 200, Kyung Hee University Hospital, KUH set). Two individual CNNs and two classification ensembles (CNNE1 and CNNE2) were tested to differentiate malignant and benign thyroid nodules. CNNs demonstrated high area under the curves (AUCs) to diagnose malignant thyroid nodules (0.898-0.937 for the internal test set and 0.821-0.885 for the external test sets). AUC was significantly higher for CNNE2 than radiologists in the SH test set (0.932 vs. 0.840, P < 0.001). AUC was not significantly different between CNNE2 and radiologists in the external test sets (P = 0.113, 0.126, and 0.690). CNN showed diagnostic performances comparable to expert radiologists for differentiating thyroid nodules on US in both the internal and external test sets.

The aim of this study was to evaluate an epidemiologic association of asthma and chronic rhinosinusitis (CRS) using a national sample cohort of the Korean population. We collected data from the Korean Health Insurance Review and Assessment Service-National Sample Cohort between 2002 and 2013, and two different case-control cohorts were designed (1st: asthmatic patients matched in a 1:1 ratio with 204,119 non-asthmatics as control I, 2nd: CRS patients matched in a 1:4 ratio with 124,020 non-CRS patients as control II). Bidirectional association was examined using Cox proportional hazard models stratified by age, sex, income, and region of residence. Patients with asthma had an increased risk of developing CRS [adjusted hazard ratio (95% confidence interval) = 1.74 (1.67-1.80)], both with nasal polyps [1.55 (1.36-1.78)], without nasal polyps [1.74 (1.67-1.81)]. In the second cohort, patients with CRS had increased risk of developing asthma [1.85 (1.80-1.91)] with similar results for those with and without nasal polyps. The strongest association for risk of CRS was in 20-39 years old men with asthma [2.41 (1.97-2.96)], while the strongest association for increased risk of asthma in those with CRS group was also seen in this same subgroup [2.40 (2.18-2.63)]. CRS and asthma had a bidirectional influence on each other. CRS increased the risk of asthma, and asthma increased the risk of CRS, especially in young men.

Humans express their emotions in various ways, such as through facial expressions and voices. In particular, emotions are directly expressed or indirectly implied in the text of utterance. Research on the technology to identify emotions included in human speech and generate utterances is being conducted in conversational artificial intelligence technology. Despite the importance of recognizing the factors of previously generated emotions to generate emotion-based utterances, most of the existing datasets only provide the classification of emotions in text and utterances. In addition, in the case of Korean datasets, the classification of emotions is not diverse, and it is mainly biased toward negative emotion classification. In this paper, we propose KEmoFact, a Korean emotion-factor dataset for extracting emotion and factors in Korean conversations. We also define two tasks for the KEmoFact dataset, EFE (Emotion Factor Extraction) and EFPE (Emotion-Factor Pair Extraction), and propose baseline models for the tasks. We contribute to the study of conversational artificial intelligence, especially in Korean, one of the low-resource languages, by proposing the KEmoFact dataset and suggesting baseline models for two tasks., (© 2023. The Author(s).)

Penetration enhancement of renewable energy sources is a core component of Korean green-island microgrid projects. This approach calls for a robust energy management system to control the stochastic behavior of renewable energy sources. Therefore, in this paper, we put forward a novel reinforcement learning-driven optimization solution for the convex problem arrangement of the Gasa island microgrid energy management as one of the prominent pilots of the Korean green islands project. We manage the convergence speed of the alternating direction method of multipliers solution for this convex problem by accurately estimating the penalty parameter with the soft actor-critic technique. However, in this arrangement, the soft actor-critic faces sparse reward hindrance, which we address here with the normalizing flow policy. Furthermore, we study the effect of demand response implementation in the Gasa island microgrid to reduce the diesel generator dependency of the microgrid and provide benefits, such as peak-shaving and gas emission reduction., (© 2022. The Author(s).)

In this study, four novel bacterial strains, USB13 T , AW1 T , GTP1 T , and HM2 T , were isolated from various environments in Busan and Jeju Island, Republic of Korea. The 16S rRNA sequencing results indicated that the four novel strains belong to the genus Ramlibacter. All four strains were tested for their potential cellulolytic properties, where strain USB13 T was identified as the only novel bacterium and the first within its genus to show cellulolytic activity. When tested, the highest activities of endoglucanase, exoglucanase, β-glucosidase, and filter paper cellulase (FPCase) were 1.91 IU/mL, 1.77 IU/mL, 0.76 IU/mL, and 1.12 IU/mL, respectively at pH 6.0. Comparisons of draft whole genome sequences (WGS) were also made using average nucleotide identity, digital DNA-DNA hybridization values, and average amino acid identity values, while whole genome comparison was visualized using the BLAST Ring Image Generator. The G + C contents of the strains ranged from 67.9 to 69.9%, while genome sizes ranged from 4.31 to 6.15 Mbp. Based on polyphasic evidence, the novel strains represent four new species within the genus Ramlibacter, for which the names Ramlibacter cellulosilyticus sp. nov. (type strain, USB13 T  = KACC 21656 T  = NBRC 114839 T ) Ramlibacter aurantiacus sp. nov. (type strain, AW1 T  = KACC 21544 T  = NBRC 114862 T ), Ramlibacter albus sp. nov. (type strain, GTP1 T  = KACC 21702 T  = NBRC 114488 T ), and Ramlibacter pallidus sp. nov. (type strain, HM2 T  = KCTC 82557 T  = NBRC 114489 T ) are proposed., (© 2022. The Author(s).)

Ground subsidence caused by natural factors, including groundwater, has been extensively researched. However, there have been few studies on ground sink caused mainly by artifacts, including underground pipelines in urban areas. This paper proposes a method of predicting ground sink susceptibility caused by underground pipelines. Underground pipeline data, drilling data, and 77 points of ground sink occurrence were collected for five 1 × 1 km urban areas. Furthermore, three ground sink conditioning factors (GSCFs) (pipe deterioration, diameter, and length) were identified by correlation analysis. Pipe deterioration showed the highest correlation with ground sink occurrence, followed by pipe length and pipe diameter in that order. Next, four machine learning methods [multinomial logistic regression (MLR), decision tree (DT) classifier, random forest (RF) classifier, and gradient boosting (GB) classifier] were applied. The results show that GB classifier had the highest accuracy of 0.7432, whereas the accuracy of RF classifier was 0.7407; thus, GB classifier was not significantly more accurate. RF classifier showed the highest reliability (0.84, 0.70, 0.87) according to the area under the receiver operating characteristic (AUC-ROC) curve. Ground sink susceptibility maps (GSSMs) of the five regions in an urban area were created using RF classifier, which performed the best overall., (© 2022. The Author(s).)

With advances in deep learning and natural language processing (NLP), the analysis of medical texts is becoming increasingly important. Nonetheless, despite the importance of processing medical texts, no research on Korean medical-specific language models has been conducted. The Korean medical text is highly difficult to analyze because of the agglutinative characteristics of the language, as well as the complex terminologies in the medical domain. To solve this problem, we collected a Korean medical corpus and used it to train the language models. In this paper, we present a Korean medical language model based on deep learning NLP. The model was trained using the pre-training framework of BERT for the medical context based on a state-of-the-art Korean language model. The pre-trained model showed increased accuracies of 0.147 and 0.148 for the masked language model with next sentence prediction. In the intrinsic evaluation, the next sentence prediction accuracy improved by 0.258, which is a remarkable enhancement. In addition, the extrinsic evaluation of Korean medical semantic textual similarity data showed a 0.046 increase in the Pearson correlation, and the evaluation for the Korean medical named entity recognition showed a 0.053 increase in the F1-score., (© 2022. The Author(s).)

Patients with weak or no symptoms accelerate the spread of COVID-19 through various mutations and require more aggressive and active means of validating the COVID-19 infection. More than 30% of patients are reported as asymptomatic infection after the delta mutation spread in Korea. It means that there is a need for a means to more actively and accurately validate the infection of the epidemic via pre-symptomatic detection, besides confirming the infection via the symptoms. Mishara et al. (Nat Biomed Eng 4, 1208-1220, 2020) reported that physiological data collected from smartwatches could be an indicator to suspect COVID-19 infection. It shows that it is possible to identify an abnormal state suspected of COVID-19 by applying an anomaly detection method for the smartwatch's physiological data and identifying the subject's abnormal state to be observed. This paper proposes to apply the One Class-Support Vector Machine (OC-SVM) for pre-symptomatic COVID-19 detection. We show that OC-SVM can provide better performance than the Mahalanobis distance-based method used by Mishara et al. (Nat Biomed Eng 4, 1208-1220, 2020) in three aspects: earlier (23.5-40% earlier) and more detection (13.2-19.1% relative better) and fewer false positives. As a result, we could conclude that OC-SVM using Resting Heart Rate (RHR) with 350 and 300 moving average size is the most recommended technique for COVID-19 pre-symptomatic detection based on physiological data from the smartwatch., (© 2022. The Author(s).)

Interspecific hybridization has been suggested to occur frequently in Rumex (Polygonaceae). Several hypothesized combinations of parental species of hybrids based on their intermediate morphology have been suggested in the genus, but few of them have been phylogenetically tested. We analyzed nuclear and chloroplast DNA sequence data of a putative natural hybrid between Rumex crispus and Rumex obtusifolius from Korea to confirm its hybrid status and to determine the maternal parent. Analysis of the nuclear DNA pgiC region revealed that R. crispus and R. obtusifolius have contributed to the nuclear genome of the putative hybrids. The haplotype distribution pattern inferred from the combined sequence data set of five chloroplast DNA regions (matK, rbcL-accD IGS, trnK-rps16 IGS, ycf6-psbM IGS and psbA-trnH IGS) indicated bidirectional hybridization events between R. crispus and R. obtusifolius. This paper provides the first molecular evidence for interspecific hybridization between R. crispus and R. obtusifolius. In addition, our findings strongly suggested that Korean populations of Rumex japonicus have a hybrid origin, and R. crispus may represent one of the parental taxa., (© 2022. The Author(s).)

Early forecasting of COVID-19 virus spread is crucial to decision making on lockdown or closure of cities, states or countries. In this paper we design a recursive bifurcation model for analyzing COVID-19 virus spread in different countries. The bifurcation facilitates recursive processing of infected population through linear least-squares fitting. In addition, a nonlinear least-squares fitting procedure is utilized to predict the future values of infected populations. Numerical results on the data from two countries (South Korea and Germany) indicate the effectiveness of our approach, compared to a logistic growth model and a Richards model in the context of early forecast. The limitation of our approach and future research are also mentioned at the end of this paper.

Although postoperative delirium is a common complication in older patients, few papers have described risk factors after of spinal surgery. The purpose of this study was to analyze various perioperative risk factors for delirium after spinal surgery in older patients. This study was performed on retrospective data collection with prospective design. We analyzed 138 patients over 65 years of age who underwent spinal surgery. Preoperative factors were cognitive function (Mini-Mental State Examination-Korean (MMSE-K) and the Korean version of the Delirium Rating Scale-Revised-98 (K-DRS 98)), age, sex, type of admission, American Society of Anesthesiologist classification, metabolic equivalents, laboratory findings, visual analog scale, and Oswestry Disability Index. Intraoperative factors were operation time, blood loss, and type of procedure. Postoperative factors were blood transfusion and type of postoperative pain control. Postoperative delirium developed in 25 patients (18.16%). Patients were divided into two groups: Group with delirium (group A) and group without delirium (group B). MMSE-K scores in Group A were significantly lower than in Group B (p < 0.001). K-DRS 98 scores were significantly higher in Group A than Group B (p < 0.001). The operation time was longer in Group A than Group B (p = 0.059). On multivariate regression analysis, the odds ratio of K-DRS 98 was 2.43 (p = 0.010). After correction for the interaction between age and MMSE-K, patients younger than 73 years old had a significantly lower incidence of delirium with higher MMSE-K score (p = 0.0014). Older age, low level of preoperative cognitive function, long duration of surgery, and transfusion were important risk factors of postoperative delirium after spinal surgery. It is important to recognize perioperative risk factors and manage appropriately.

Forest succession from Pinus to Quercus has often been observed in temperate forest, although the succession mechanism is not clearly understood. This study investigated factors that affect the succession of forests from pine to oak, using forest vegetation inventory data at plots at Kwan-ak mountain in Korea. Analyses of understory canopy coverage, light intensity, and tree numbers and ages in P. densiflora forests indicate that Q. mongolica can only invade these forests before understory shrub establishment. The results from analyses of all environmental factors indicate that similar adverse effects from environmental factors occur in established P. densiflora and Q. mongolica forests that inhibit Q. mongolica seedling survival. However, the observed survival rate of Q. mongolica seedlings under P. densiflora during winter were much higher than Q. mongolica seedlings under Q. mongolica trees, and it is due to accumulated snow over Q. mongolica forest litter which breaks or inhibits the emergence of Q. mongolica seedlings. Protecting seedlings with plastic cups significantly increased the survival rate which confirms that forest floor litter acts as a filter for the regeneration and succession of Q. mongolica forests. This paper thus concludes that understory shrubs and forest litter affect the succession dynamics of P. densiflora and Q. mongolica forests.

Mean birth weight of twins is known to be lower than that of singletons, however, southern China lacks a twin-specific birth weight reference. In this paper, we use data from the Birth Certificate System in southern China, collected between January 1 st 2014 and December 31 st 2017 and including 161,076 twins, to calculate sex- and gestational week-specific birth weight percentiles (the 3 rd , 10 th , 25 th , 50 th , 75 th , 90 th , and 97 th ). We applied generalized additive models for location, scale and shape (GAMLSS) when calculating the birth weight percentiles, and calculated percentiles for monochorionic and dichorionic twins separately. We next used data collected between Jan 1st 2018 and Apr 30th 2018, encompassing 12,371 live births, to calculate the SGA and LGA ratios using birth weight references in Australia, South Korea and China (based on birth defects surveillance system) and birth weight percentiles calculated in this study. Compared to dichorionic twins, monochorionic twins had lower birth weights at 25 to 42 weeks of gestation. The calculated SGA and LGA ratios were relatively stable compared to the other references.