1.Occupation classification model based on DistilKoBERT: using the 5th and 6th Korean Working Condition Surveys
Tae-Yeon KIM ; Seong-Uk BAEK ; Myeong-Hun LIM ; Byungyoon YUN ; Domyung PAEK ; Kyung Ehi ZOH ; Kanwoo YOUN ; Yun Keun LEE ; Yangho KIM ; Jungwon KIM ; Eunsuk CHOI ; Mo-Yeol KANG ; YoonHo CHO ; Kyung-Eun LEE ; Juho SIM ; Juyeon OH ; Heejoo PARK ; Jian LEE ; Jong-Uk WON ; Yu-Min LEE ; Jin-Ha YOON
Annals of Occupational and Environmental Medicine 2024;36(1):e19-
Accurate occupation classification is essential in various fields, including policy development and epidemiological studies. This study aims to develop an occupation classification model based on DistilKoBERT. This study used data from the 5th and 6th Korean Working Conditions Surveys conducted in 2017 and 2020, respectively. A total of 99,665 survey participants, who were nationally representative of Korean workers, were included. We used natural language responses regarding their job responsibilities and occupational codes based on the Korean Standard Classification of Occupations (7th version, 3-digit codes). The dataset was randomly split into training and test datasets in a ratio of 7:3. The occupation classification model based on DistilKoBERT was fine-tuned using the training dataset, and the model was evaluated using the test dataset. The accuracy, precision, recall, and F1 score were calculated as evaluation metrics. The final model, which classified 28,996 survey participants in the test dataset into 142 occupational codes, exhibited an accuracy of 84.44%. For the evaluation metrics, the precision, recall, and F1 score of the model, calculated by weighting based on the sample size, were 0.83, 0.84, and 0.83, respectively. The model demonstrated high precision in the classification of service and sales workers yet exhibited low precision in the classification of managers. In addition, it displayed high precision in classifying occupations prominently represented in the training dataset. This study developed an occupation classification system based on DistilKoBERT, which demonstrated reasonable performance. Despite further efforts to enhance the classification accuracy, this automated occupation classification model holds promise for advancing epidemiological studies in the fields of occupational safety and health.
2.The Third Nationwide Korean Heart Failure III Registry (KorHF III):The Study Design Paper
Minjae YOON ; Eung Ju KIM ; Seong Woo HAN ; Seong-Mi PARK ; In-Cheol KIM ; Myeong-Chan CHO ; Hyo-Suk AHN ; Mi-Seung SHIN ; Seok Jae HWANG ; Jin-Ok JEONG ; Dong Heon YANG ; Jae-Joong KIM ; Jin Oh CHOI ; Hyun-Jai CHO ; Byung-Su YOO ; Seok-Min KANG ; Dong-Ju CHOI
International Journal of Heart Failure 2024;6(2):70-75
With advancements in both pharmacologic and non-pharmacologic treatments, significant changes have occurred in heart failure (HF) management. The previous Korean HF registries, namely the Korea Heart Failure Registry (KorHF-registry) and Korean Acute Heart Failure Registry (KorAHF-registry), no longer accurately reflect contemporary acute heart failure (AHF) patients. Our objective is to assess contemporary AHF patients through a nationwide registry encompassing various aspects, such as clinical characteristics, management approaches, hospital course, and long-term outcomes of individuals hospitalized for AHF in Korea. This prospective observational multicenter cohort study (KorHF III) is organized by the Korean Society of Heart Failure. We aim to prospectively enroll 7,000 or more patients hospitalized for AHF at 47 tertiary hospitals in Korea starting from March 2018. Eligible patients exhibit signs and symptoms of HF and demonstrate either lung congestion or objective evidence of structural or functional cardiac abnormalities in echocardiography, or isolated right-sided HF. Patients will be followed up for up to 5 years after enrollment in the registry to evaluate long-term clinical outcomes. KorHF III represents the nationwide AHF registry that will elucidate the clinical characteristics, management strategies, and outcomes of contemporary AHF patients in Korea.
3.Occupation classification model based on DistilKoBERT: using the 5th and 6th Korean Working Condition Surveys
Tae-Yeon KIM ; Seong-Uk BAEK ; Myeong-Hun LIM ; Byungyoon YUN ; Domyung PAEK ; Kyung Ehi ZOH ; Kanwoo YOUN ; Yun Keun LEE ; Yangho KIM ; Jungwon KIM ; Eunsuk CHOI ; Mo-Yeol KANG ; YoonHo CHO ; Kyung-Eun LEE ; Juho SIM ; Juyeon OH ; Heejoo PARK ; Jian LEE ; Jong-Uk WON ; Yu-Min LEE ; Jin-Ha YOON
Annals of Occupational and Environmental Medicine 2024;36(1):e19-
Accurate occupation classification is essential in various fields, including policy development and epidemiological studies. This study aims to develop an occupation classification model based on DistilKoBERT. This study used data from the 5th and 6th Korean Working Conditions Surveys conducted in 2017 and 2020, respectively. A total of 99,665 survey participants, who were nationally representative of Korean workers, were included. We used natural language responses regarding their job responsibilities and occupational codes based on the Korean Standard Classification of Occupations (7th version, 3-digit codes). The dataset was randomly split into training and test datasets in a ratio of 7:3. The occupation classification model based on DistilKoBERT was fine-tuned using the training dataset, and the model was evaluated using the test dataset. The accuracy, precision, recall, and F1 score were calculated as evaluation metrics. The final model, which classified 28,996 survey participants in the test dataset into 142 occupational codes, exhibited an accuracy of 84.44%. For the evaluation metrics, the precision, recall, and F1 score of the model, calculated by weighting based on the sample size, were 0.83, 0.84, and 0.83, respectively. The model demonstrated high precision in the classification of service and sales workers yet exhibited low precision in the classification of managers. In addition, it displayed high precision in classifying occupations prominently represented in the training dataset. This study developed an occupation classification system based on DistilKoBERT, which demonstrated reasonable performance. Despite further efforts to enhance the classification accuracy, this automated occupation classification model holds promise for advancing epidemiological studies in the fields of occupational safety and health.
4.Occupation classification model based on DistilKoBERT: using the 5th and 6th Korean Working Condition Surveys
Tae-Yeon KIM ; Seong-Uk BAEK ; Myeong-Hun LIM ; Byungyoon YUN ; Domyung PAEK ; Kyung Ehi ZOH ; Kanwoo YOUN ; Yun Keun LEE ; Yangho KIM ; Jungwon KIM ; Eunsuk CHOI ; Mo-Yeol KANG ; YoonHo CHO ; Kyung-Eun LEE ; Juho SIM ; Juyeon OH ; Heejoo PARK ; Jian LEE ; Jong-Uk WON ; Yu-Min LEE ; Jin-Ha YOON
Annals of Occupational and Environmental Medicine 2024;36(1):e19-
Accurate occupation classification is essential in various fields, including policy development and epidemiological studies. This study aims to develop an occupation classification model based on DistilKoBERT. This study used data from the 5th and 6th Korean Working Conditions Surveys conducted in 2017 and 2020, respectively. A total of 99,665 survey participants, who were nationally representative of Korean workers, were included. We used natural language responses regarding their job responsibilities and occupational codes based on the Korean Standard Classification of Occupations (7th version, 3-digit codes). The dataset was randomly split into training and test datasets in a ratio of 7:3. The occupation classification model based on DistilKoBERT was fine-tuned using the training dataset, and the model was evaluated using the test dataset. The accuracy, precision, recall, and F1 score were calculated as evaluation metrics. The final model, which classified 28,996 survey participants in the test dataset into 142 occupational codes, exhibited an accuracy of 84.44%. For the evaluation metrics, the precision, recall, and F1 score of the model, calculated by weighting based on the sample size, were 0.83, 0.84, and 0.83, respectively. The model demonstrated high precision in the classification of service and sales workers yet exhibited low precision in the classification of managers. In addition, it displayed high precision in classifying occupations prominently represented in the training dataset. This study developed an occupation classification system based on DistilKoBERT, which demonstrated reasonable performance. Despite further efforts to enhance the classification accuracy, this automated occupation classification model holds promise for advancing epidemiological studies in the fields of occupational safety and health.
5.The Real-World Outcome of First Line Atezolizumab in Extensive-Stage Small Cell Lung Cancer: A Multicenter Prospective Cohort Study
Myeong Geun CHOI ; Yeon Joo KIM ; Jae Cheol LEE ; Wonjun JI ; In-Jae OH ; Sung Yong LEE ; Seong Hoon YOON ; Shin Yup LEE ; Jeong Eun LEE ; Eun Young KIM ; Chang-Min CHOI
Cancer Research and Treatment 2024;56(2):422-429
Purpose:
The addition of immune checkpoint inhibitors to chemotherapy has improved survival outcomes in patients with extensive-stage small cell lung cancer (ES-SCLC). However, their real-world effectiveness remains unknown. Therefore, we investigated the effectiveness of atezolizumab plus chemotherapy in ES-SCLC in actual clinical settings.
Materials and Methods:
In this multicenter prospective cohort study, patients with ES-SCLC receiving or scheduled to receive atezolizumab in combination with etoposide and carboplatin were enrolled between June 2021 and August 2022. The primary outcomes were progression-free survival (PFS) and the 1-year overall survival (OS) rate.
Results:
A total of 100 patients with ES-SCLC were enrolled from seven centers. Median age was 69 years, and 6% had an Eastern Cooperative Oncology Group performance status (ECOG PS) ≥ 2. The median PFS was 6.0 months, the 1-year OS rate was 62.2%, and the median OS was 13.5 months. An ECOG PS of 2-3 and progressive disease as the best response were poor prognostic factors for PFS, while an ECOG PS of 2-3 and brain metastasis were associated with poor prognosis for OS. In addition, consolidative thoracic radiotherapy was found to be an independent favorable prognostic factor for OS (hazard ratio, 0.336; p=0.021). Grade ≥ 3 treatment-related adverse events were observed in 7% of patients, with treatment-related deaths occurring in 2% of patients.
Conclusion
We provided evidence of the favorable real-world effectiveness and safety of atezolizumab plus chemotherapy in ES-SCLC patients, including in the elderly and those with poor ECOG PS. Additional consolidative thoracic radiotherapy may also benefit ES-SCLC patients.
6.Association between serum osteoprotegerin level and renal prognosis in nondialysis patients with chronic kidney disease in the Korean Cohort Study for Outcomes in Patients with Chronic Kidney Disease (the KNOW-CKD Study)
Tae Ryom OH ; Chana MYEONG ; Su Hyun SONG ; Hong Sang CHOI ; Sang Heon SUH ; Chang Seong KIM ; Eun Hui BAE ; Wookyung CHUNG ; Kyu Hun CHOI ; Kook Hwan OH ; Seong Kwon MA ; Soo Wan KIM
Kidney Research and Clinical Practice 2022;41(2):200-208
Osteoprotegerin is an important regulator of bone metabolism and vascular calcification. The association between serum osteoprotegerin level and chronic kidney disease (CKD) progression has not been elucidated. We investigated the prognostic value of serum osteoprotegerin levels in nondialysis CKD patients. Methods: We analyzed 2,082 patients enrolled in the Korean Cohort Study for Outcomes in Patients with CKD between 2011 and 2016. Patients were divided into quartiles by their serum osteoprotegerin levels. The primary outcome was the occurrence of ≥1 of the following: dialysis initiation, kidney transplantation, a two-fold increase in serum creatinine level from baseline, or a 50% decrease in the estimated glomerular filtration rate (eGFR). Cox proportional hazard regression models were used to investigate the prognostic value of the serum osteoprotegerin level to CKD progression. Results: The median follow-up period was 48.9 months, and 641 patients (30.8%) experienced the primary outcome. The hazard ratio of serum osteoprotegerin for renal progression in the full extended Cox proportional hazard model was 1.064 (95% confidence interval, 1.041–1.088). Subgroup analyses by age, presence of diabetes, and eGFR showed significant results consistent with the overall analysis results. Conclusion: Serum osteoprotegerin level is independently associated with renal prognosis and could have prognostic importance in CKD progression.
7.The effect of probiotic supplementation on systemic inflammation in dialysis patients
Eunho CHOI ; Jihyun YANG ; Geun-Eog JI ; Myeong Soo PARK ; Yeongje SEONG ; Se Won OH ; Myung Gyu KIM ; Won Yong CHO ; Sang Kyung JO
Kidney Research and Clinical Practice 2022;41(1):89-101
Emerging evidence suggests that intestinal dysbiosis contributes to systemic inflammation and cardiovascular diseases in dialysis patients. The purpose of this study was to evaluate the effects of probiotic supplementation on various inflammatory parameters in hemodialysis (HD) patients. Methods: Twenty-two patients with maintenance HD were enrolled. These patients were treated twice a day with 2.0 ×1010 colony forming units of a combination of Bifidobacterium bifidum BGN4 and Bifidobacterium longum BORI for 3 months. The microbiome and fecal short-chain fatty acids (SCFAs) were analyzed. The percentages of CD14+ CD16+ proinflammatory monocytes and CD4+ CD25+ regulatory T-cells (Tregs) before and after probiotic supplementation were determined by flow cytometry. Serum levels of calprotectin and cytokine responses upon lipopolysaccharide (LPS) challenge were compared before and after probiotic supplementation. Results: Fecal SCFAs increased significantly after probiotic supplementation. Serum levels of calprotectin and interleukin 6 upon LPS stimulation significantly decreased. The anti-inflammatory effects of probiotics were associated with a significant increase in the percentage of CD4+ CD25+ Tregs (3.5% vs. 8.6%, p < 0.05) and also with a decrease of CD14+ CD16+ proinflammatory monocytes (310/ mm2 vs. 194/mm2 , p < 0.05). Conclusion: Probiotic supplementation reduced systemic inflammatory responses in HD patients and this effect was associated with an increase in Tregs and a decrease in proinflammatory monocytes. Hence, targeting intestinal dysbiosis might be a novel strategy for decreasing inflammation and cardiovascular risks in HD patients.
8.The current status and outcomes of in-hospital P2Y12 receptor inhibitor switching in Korean patients with acute myocardial infarction
Keun-Ho PARK ; Myung Ho JEONG ; Hyun Kuk KIM ; Young-Jae KI ; Sung Soo KIM ; Youngkeun AHN ; Hyun Yi KOOK ; Hyo-Soo KIM ; Hyeon Cheol GWON ; Ki Bae SEUNG ; Seung Woon RHA ; Shung Chull CHAE ; Chong Jin KIM ; Kwang Soo CHA ; Jong Seon PARK ; Jung Han YOON ; Jei Keon CHAE ; Seung Jae JOO ; Dong-Joo CHOI ; Seung Ho HUR ; In Whan SEONG ; Myeong Chan CHO ; Doo Il KIM ; Seok Kyu OH ; Tae Hoon AHN ; Jin Yong HWANG ;
The Korean Journal of Internal Medicine 2022;37(2):350-365
Background/Aims:
While switching strategies of P2Y12 receptor inhibitors (RIs) have sometimes been used in acute myocardial infarction (AMI) patients, the current status of in-hospital P2Y12RI switching remains unknown.
Methods:
Overall, 8,476 AMI patients who underwent successful revascularization from Korea Acute Myocardial Infarction Registry-National Institute of Health (KAMIR-NIH) were divided according to in-hospital P2Y12RI strategies, and net adverse cardiovascular events (NACEs), defined as a composite of cardiac death, non-fatal myocardial infarction (MI), stroke, or thrombolysis in myocardial infarction (TIMI) major bleeding during hospitalization were compared.
Results:
Patients with in-hospital P2Y12RI switching accounted for 16.5%, of which 867 patients were switched from clopidogrel to potent P2Y12RI (C-P) and 532 patients from potent P2Y12RI to clopidogrel (P-C). There were no differences in NACEs among the unchanged clopidogrel, the unchanged potent P2Y12RIs, and the P2Y12RI switching groups. However, compared to the unchanged clopidogrel group, the C-P group had a higher incidence of non-fatal MI, and the P-C group had a higher incidence of TIMI major bleeding. In clinical events of in-hospital P2Y12RI switching, 90.9% of non-fatal MI occurred during pre-switching clopidogrel administration, 60.7% of TIMI major bleeding was related to pre-switching P2Y12RIs, and 71.4% of TIMI major bleeding was related to potent P2Y12RIs. Only 21.6% of the P2Y12RI switching group switched to P2Y12RIs after a loading dose (LD); however, there were no differences in clinical events between patients with and without LD.
Conclusions
In-hospital P2Y12RI switching occurred occasionally, but had relatively similar clinical outcomes compared to unchanged P2Y12RIs in Korean AMI patients. Non-fatal MI and bleeding appeared to be mainly related to pre-switching P2Y12RIs.
9.Effects of Virtual Reality-Based Cognitive Training in the Elderly with and without Mild Cognitive Impairment
Seri MAENG ; Jin Pyo HONG ; Won-Hyoung KIM ; Hyeyoung KIM ; Seo-Eun CHO ; Jae Myeong KANG ; Kyoung-Sae NA ; Seok-Hee OH ; Jung Woon PARK ; Jae Nam BAE ; Seong-Jin CHO
Psychiatry Investigation 2021;18(7):619-627
Objective:
This study aimed to introduce a 4-week long fully immersive virtual reality-based cognitive training (VRCT) program that could be applied for both a cognitively normal elderly population and patients with mild cognitive impairment (MCI). In addition, we attempted to investigate the neuropsychological effects of the VRCT program in each group.
Methods:
A total of 56 participants, 31 in the MCI group and 25 in the cognitively normal elderly group, underwent eight sessions of VRCT for 4 weeks. In order to evaluate the effects of the VRCT, the Korean version of the Consortium to Establish a Registry for Alzheimer’s Disease Assessment Packet was administered before and after the program. The program’ s safety was assessed using a simulator sickness questionnaire (SSQ), and availability was assessed using the presence questionnaire.
Results:
After the eighth session of the VRCT program, cognitive improvement was observed in the ability to learn new information, visuospatial constructional ability, and frontal lobe function in both groups. At the baseline evaluation, based on the SSQ, the MCI group complained of disorientation and nausea significantly more than the cognitively normal elderly group did. However, both groups showed a reduction in discomfort as the VRCT program progressed.
Conclusion
We conclude that our VRCT program helps improve cognition in both the MCI group and cognitively normal elderly group. Therefore, the VRCT is expected to help improve cognitive function in elderly populations with and without MCI.
10.Relationship Between Sleep Duration and Suicidal Ideation Among Farmers: Safety for Agricultural Injuries of Farmers Cohort Study of Jeju, Korea
Jung-Hwan OH ; Jeong Rae YOO ; Seo Young KO ; Jeong Ho KANG ; Sung Kgun LEE ; Wooseong JEONG ; Gil Myeong SEONG ; Hyun Jung LEE ; Sung Wook SONG
Safety and Health at Work 2021;12(1):102-107
Background:
This study was conducted to identify the association between sleep duration and suicidal ideation among farmers in Korea.
Methods:
We used Safety for Agricultural Injuries of Farmers cohort data collected from September 2015 to June 2018, which was an agricultural survey on the health and behaviors of adult farmers in Jeju island, Korea.
Results:
A total of 964 participants were included in the analysis, and 3.7% of them were identified with having suicidal ideation. The frequencies of average daily sleep duration of <6 h (short sleep), 6–8 h (normal sleep), and >8 h (long sleep) were 24.4%, 70%, and 5.6%, respectively. Multivariate analyses revealed that short sleep duration was significantly associated with suicidal ideation compared with normal sleep duration (odds ratio = 2.49, 95% confidence interval = 1.07–5.77).
Conclusion
Our findings suggest that short sleep duration in farmers result in higher suicidal ideation. Because individuals who have suicidal ideation often commit suicide, careful monitoring is required to prevent suicide in farmers with short sleep duration.

Result Analysis
Print
Save
E-mail