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.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.
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.Lazertinib versus Gefitinib as First-Line Treatment for EGFR-mutated Locally Advanced or Metastatic NSCLC: LASER301 Korean Subset
Ki Hyeong LEE ; Byoung Chul CHO ; Myung-Ju AHN ; Yun-Gyoo LEE ; Youngjoo LEE ; Jong-Seok LEE ; Joo-Hang KIM ; Young Joo MIN ; Gyeong-Won LEE ; Sung Sook LEE ; Kyung-Hee LEE ; Yoon Ho KO ; Byoung Yong SHIM ; Sang-We KIM ; Sang Won SHIN ; Jin-Hyuk CHOI ; Dong-Wan KIM ; Eun Kyung CHO ; Keon Uk PARK ; Jin-Soo KIM ; Sang Hoon CHUN ; Jangyoung WANG ; SeokYoung CHOI ; Jin Hyoung KANG
Cancer Research and Treatment 2024;56(1):48-60
Purpose:
This subgroup analysis of the Korean subset of patients in the phase 3 LASER301 trial evaluated the efficacy and safety of lazertinib versus gefitinib as first-line therapy for epidermal growth factor receptor mutated (EGFRm) non–small cell lung cancer (NSCLC).
Materials and Methods:
Patients with locally advanced or metastatic EGFRm NSCLC were randomized 1:1 to lazertinib (240 mg/day) or gefitinib (250 mg/day). The primary endpoint was investigator-assessed progression-free survival (PFS).
Results:
In total, 172 Korean patients were enrolled (lazertinib, n=87; gefitinib, n=85). Baseline characteristics were balanced between the treatment groups. One-third of patients had brain metastases (BM) at baseline. Median PFS was 20.8 months (95% confidence interval [CI], 16.7 to 26.1) for lazertinib and 9.6 months (95% CI, 8.2 to 12.3) for gefitinib (hazard ratio [HR], 0.41; 95% CI, 0.28 to 0.60). This was supported by PFS analysis based on blinded independent central review. Significant PFS benefit with lazertinib was consistently observed across predefined subgroups, including patients with BM (HR, 0.28; 95% CI, 0.15 to 0.53) and those with L858R mutations (HR, 0.36; 95% CI, 0.20 to 0.63). Lazertinib safety data were consistent with its previously reported safety profile. Common adverse events (AEs) in both groups included rash, pruritus, and diarrhoea. Numerically fewer severe AEs and severe treatment–related AEs occurred with lazertinib than gefitinib.
Conclusion
Consistent with results for the overall LASER301 population, this analysis showed significant PFS benefit with lazertinib versus gefitinib with comparable safety in Korean patients with untreated EGFRm NSCLC, supporting lazertinib as a new potential treatment option for this patient population.
5.Clinical Outcome after Everolimus-Eluting Stent Implantation for Small Vessel Coronary Artery Disease: XIENCE Asia Small Vessel Study
Doo Sun SIM ; Dae Young HYUN ; Young Joon HONG ; Ju Han KIM ; Youngkeun AHN ; Myung Ho JEONG ; Sang Rok LEE ; Jei Keon CHAE ; Keun Ho PARK ; Young Youp KOH ; Kyeong Ho YUN ; Seok Kyu OH ; Seung Jae JOO ; Sun Ho HWANG ; Jong Pil PARK ; Jay Young RHEW ; Su Hyun KIM ; Jang Hyun CHO ; Seung Uk LEE ; Dong Goo KANG
Chonnam Medical Journal 2024;60(1):78-86
There are limited data on outcomes after implantation of everolimus-eluting stents (EES) in East Asian patients with small vessel coronary lesions. A total of 1,600 patients treated with XIENCE EES (Abbott Vascular, CA, USA) were divided into the small vessel group treated with one ≤2.5 mm stent (n=119) and the non-small vessel group treated with one ≥2.75 mm stent (n=933). The primary end point was a patient-oriented composite outcome (POCO), a composite of all-cause death, myocardial infarction (MI), and any repeat revascularization at 12 months. The key secondary end point was a device-oriented composite outcome (DOCO), a composite of cardiovascular death, target-vessel MI, and target lesion revascularization at 12 months. The small vessel group was more often female, hypertensive, less likely to present with ST-elevation MI, and more often treated for the left circumflex artery, whereas the non-small vessel group more often had type B2/C lesions, underwent intravascular ultrasound, and received unfractionated heparin. In the propensity matched cohort, the mean stent diameter was 2.5±0.0 mm and 3.1±0.4 mm in the small and non-small vessel groups, respectively. Propensity-adjusted POCO at 12 months was 6.0% in the small vessel group and 4.3% in the non-small vessel group (p=0.558). There was no significant difference in DOCO at 12 months (small vessel group: 4.3% and non-small vessel group: 1.7%, p=0.270).Outcomes of XIENCE EES for small vessel disease were comparable to those for non-small vessel disease at 12-month clinical follow-up in real-world Korean patients.
6.S100 Calcium-Binding Protein A9, a Potential Novel Diagnostic Biomarker for Idiopathic Pulmonary Fibrosis
Jong-Uk LEE ; Min Kyung KIM ; Myung-Shin KIM ; Sun Ju LEE ; Seung-lee PARK ; Hun Soo CHANG ; Jong-Sook PARK ; Choon-Sik PARK
Journal of Korean Medical Science 2024;39(1):e13-
Background:
Neutrophilic inflammation is a characteristic feature of idiopathic pulmonary fibrosis (IPF). S100 calcium-binding protein A9 (S100A9) is a neutrophil-derived protein involved in the development of neutrophil-related chronic inflammatory disorders. However, the role of S100A9 in IPF remains unclear.
Methods:
We used enzyme-linked immunosorbent assays to measure S100A9 levels in bronchoalveolar lavage fluid (BALF) and serum obtained from healthy controls (HCs) and patients with IPF, non-specific interstitial pneumonia, hypersensitivity pneumonitis, and sarcoidosis.
Results:
Compared with HCs, BALF S100A9 levels were significantly higher in IPF patients (P < 0.001), patients with hypersensitivity pneumonitis (P = 0.043), and patients with nonspecific interstitial pneumonia (P < 0.001). The S100A9 level in BALF of 0.093 ng/mL could distinguish IPF patients from HCs, with a specificity of 78.8% and a sensitivity of 81.6%. Similarly, the S100A9 level in BALF of 0.239 ng/mL had a specificity of 64.7% and a sensitivity of 66.7% for distinguishing IPF patients from patients with other interstitial lung diseases. Additionally, BALF S100A9 levels were significantly correlated with neutrophil counts (r = 0.356, P < 0.001) in BALF. IPF patients with S100A9 levels in BALF > 0.533 ng/ mL had lower survival rates, compared with patients who had levels ≤ 0.553 ng/mL (n = 49; hazard ratio [HR], 3.62; P = 0.021). Combination analysis revealed that IPF patients with S100A9 levels in BALF> 0.553 ng/mL or neutrophil percentages > 49.1% (n = 43) had significantly lower survival rates than patients with S100A9 levels in BALF ≤ 0.553 ng/mL and neutrophil percentages ≤ 49.1% (n = 41) (HR, 3.91; P = 0.014). Additionally, patients with serum S100A9 levels > 0.077 ng/mL (n = 29) had significantly lower survival rates than patients with levels ≤ 0.077 ng/mL (n = 53, HR, 2.52; P = 0.013). S100A9 was expressed on neutrophils and macrophages in BALF from IPF patients as well as α-smooth muscle actin positive cells in the lung tissues.
Conclusion
S100A9 is involved in the development and progression of IPF. Moreover, S100A9 levels in BALF and serum may be surrogate markers for IPF diagnosis and survival prediction, particularly when analyzed in combination with neutrophil percentages.
7.Evaluating the Validity and Reliability of the Korean Version of the Scales for Outcomes in Parkinson’s Disease–Cognition
Jinse PARK ; Eungseok OH ; Seong-Beom KOH ; In-Uk SONG ; Tae-Beom AHN ; Sang Jin KIM ; Sang-Myung CHEON ; Yoon-Joong KIM ; Jin Whan CHO ; Hyeo-Il MA ; Mee Young PARK ; Jong Sam BAIK ; Phil Hyu LEE ; Sun Ju CHUNG ; Jong-Min KIM ; Han-Joon KIM ; Young-Hee SUNG ; Do Young KWON ; Jae-Hyeok LEE ; Jee-Young LEE ; Ji Seon KIM ; Ji Young YUN ; Hee Jin KIM ; Jin Yong HONG ; Mi-Jung KIM ; Jinyoung YOUN ; Hui-Jun YANG ; Won Tae YOON ; Sooyeoun YOU ; Kyum-Yil KWON ; Su-Yun LEE ; Younsoo KIM ; Hee-Tae KIM ; Joong-Seok KIM ; Ji-Young KIM
Journal of Movement Disorders 2024;17(3):328-332
Objective:
The Scales for Outcomes in Parkinson’s Disease–Cognition (SCOPA-Cog) was developed to assess cognition in patients with Parkinson’s disease (PD). In this study, we aimed to evaluate the validity and reliability of the Korean version of the SCOPACog (K-SCOPA-Cog).
Methods:
We enrolled 129 PD patients with movement disorders from 31 clinics in South Korea. The original version of the SCOPA-Cog was translated into Korean using the translation-retranslation method. The test–retest method with an intraclass correlation coefficient (ICC) and Cronbach’s alpha coefficient were used to assess reliability. Spearman’s rank correlation analysis with the Montreal Cognitive Assessment-Korean version (MOCA-K) and the Korean Mini-Mental State Examination (K-MMSE) were used to assess concurrent validity.
Results:
The Cronbach’s alpha coefficient was 0.797, and the ICC was 0.887. Spearman’s rank correlation analysis revealed a significant correlation with the K-MMSE and MOCA-K scores (r = 0.546 and r = 0.683, respectively).
Conclusion
Our results demonstrate that the K-SCOPA-Cog has good reliability and validity.
8.Korean clinical practice guidelines for diagnostic and procedural sedation
Sang-Hyun KIM ; Young-Jin MOON ; Min Suk CHAE ; Yea-Ji LEE ; Myong-Hwan KARM ; Eun-Young JOO ; Jeong-Jin MIN ; Bon-Nyeo KOO ; Jeong-Hyun CHOI ; Jin-Young HWANG ; Yeonmi YANG ; Min A KWON ; Hyun Jung KOH ; Jong Yeop KIM ; Sun Young PARK ; Hyunjee KIM ; Yang-Hoon CHUNG ; Na Young KIM ; Sung Uk CHOI
Korean Journal of Anesthesiology 2024;77(1):5-30
Safe and effective sedation depends on various factors, such as the choice of sedatives, sedation techniques used, experience of the sedation provider, degree of sedation-related education and training, equipment and healthcare worker availability, the patient’s underlying diseases, and the procedure being performed. The purpose of these evidence-based multidisciplinary clinical practice guidelines is to ensure the safety and efficacy of sedation, thereby contributing to patient safety and ultimately improving public health. These clinical practice guidelines comprise 15 key questions covering various topics related to the following: the sedation providers; medications and equipment available; appropriate patient selection; anesthesiologist referrals for high-risk patients; pre-sedation fasting; comparison of representative drugs used in adult and pediatric patients; respiratory system, cardiovascular system, and sedation depth monitoring during sedation; management of respiratory complications during pediatric sedation; and discharge criteria. The recommendations in these clinical practice guidelines were systematically developed to assist providers and patients in sedation-related decision making for diagnostic and therapeutic examinations or procedures. Depending on the characteristics of primary, secondary, and tertiary care institutions as well as the clinical needs and limitations, sedation providers at each medical institution may choose to apply the recommendations as they are, modify them appropriately, or reject them completely.
9.Efficacy and Safety of Lurasidone vs. Quetiapine XR in Acutely Psychotic Patients With Schizophrenia in Korea: A Randomized, Double-Blind, Active-Controlled Trial
Se Hyun KIM ; Do-Un JUNG ; Do Hoon KIM ; Jung Sik LEE ; Kyoung-Uk LEE ; Seunghee WON ; Bong Ju LEE ; Sung-Gon KIM ; Sungwon ROH ; Jong-Ik PARK ; Minah KIM ; Sung Won JUNG ; Hong Seok OH ; Han-yong JUNG ; Sang Hoon KIM ; Hyun Seung CHEE ; Jong-Woo PAIK ; Kyu Young LEE ; Soo In KIM ; Seung-Hwan LEE ; Eun-Jin CHEON ; Hye-Geum KIM ; Heon-Jeong LEE ; In Won CHUNG ; Joonho CHOI ; Min-Hyuk KIM ; Seong-Jin CHO ; HyunChul YOUN ; Jhin-Goo CHANG ; Hoo Rim SONG ; Euitae KIM ; Won-Hyoung KIM ; Chul Eung KIM ; Doo-Heum PARK ; Byung-Ook LEE ; Jungsun LEE ; Seung-Yup LEE ; Nuree KANG ; Hee Yeon JUNG
Psychiatry Investigation 2024;21(7):762-771
Objective:
This study was performed to evaluate the efficacy and safety of lurasidone (160 mg/day) compared to quetiapine XR (QXR; 600 mg/day) in the treatment of acutely psychotic patients with schizophrenia.
Methods:
Patients were randomly assigned to 6 weeks of double-blind treatment with lurasidone 160 mg/day (n=105) or QXR 600 mg/day (n=105). Primary efficacy measure was the change from baseline to week 6 in Positive and Negative Syndrome Scale (PANSS) total score and Clinical Global Impressions severity (CGI-S) score. Adverse events, body measurements, and laboratory parameters were assessed.
Results:
Lurasidone demonstrated non-inferiority to QXR on the PANSS total score. Adjusted mean±standard error change at week 6 on the PANSS total score was -26.42±2.02 and -27.33±2.01 in the lurasidone and QXR group, respectively. The mean difference score was -0.91 (95% confidence interval -6.35–4.53). The lurasidone group showed a greater reduction in PANSS total and negative subscale on week 1 and a greater reduction in end-point CGI-S score compared to the QXR group. Body weight, body mass index, and waist circumference in the lurasidone group were reduced, with significantly lower mean change compared to QXR. Endpoint changes in glucose, cholesterol, triglycerides, and low-density lipoprotein levels were also significantly lower. The most common adverse drug reactions with lurasidone were akathisia and nausea.
Conclusion
Lurasidone 160 mg/day was found to be non-inferior to QXR 600 mg/day in the treatment of schizophrenia with comparable efficacy and tolerability. Adverse effects of lurasidone were generally tolerable, and beneficial effects on metabolic parameters can be expected.
10.Evaluation of Sleep Disturbance in Alopecia Areata through Questionnaire: Pittsburgh Sleep Quality Index as a Reasonable Tool
So Hee PARK ; Ki Hwan JI ; Jong Uk KIM ; Seung Hee JANG ; Sang Woo AHN ; Seong Min HONG ; Woo Jung JIN ; Jung Eun SEOL ; Hyojin KIM
Annals of Dermatology 2023;35(3):183-189
Background:
Alopecia areata (AA) is common non-scarring hair loss disease. Sleep distrubance has been regarded as a triggering or aggravating factor for AA. However, objective evaluation of sleep disturbance and its clinical effect on AA has not been clearly demonstrated.
Objective:
This study investigated objective sleep evaluation tool for AA patients and their clinical correlation.
Methods:
Patients presenting with new-onset AA or recurrences of pre-existing AA were included, and those who reported sleep disturbance in the preliminary survey were designated as the sleep disturbance group (SD group). Sleep quality was investigated for them using three self-administered questionnaires: Pittsburgh Sleep Quality Index (PSQI), Insomnia Severity Index (ISI), and Epworth Sleep Scale (ESS). Demographic information and clinical features of AA were analyzed according to sleep quality.
Results:
A total of 400 participants were enrolled, and 53 were categorized into the SD group. The incidence of stressful events was significantly higher in the SD group (54.7%) than in the non-SD group (25.1%) (p<0.001). Based on the PSQI, 77.3% of participants were objective poor sleepers (score of 5 or more), and they showed a significantly higher incidence of stressful events compared to good sleepers (p=0.019). The proportion of poor sleepers was significantly lower in patients with mild AA (S1) than in those with moderate to severe AA (S2~S5) (p=0.045).
Conclusion
This study demonstrated a positive correlation among stress, SD, and AA. The degree of SD was objectively represented by the PSQI score, showing different scores according to AA severity.

Result Analysis
Print
Save
E-mail