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.Omission of Breast Surgery in Predicted Pathologic Complete Response after Neoadjuvant Systemic Therapy: A Multicenter, Single-Arm, Non-inferiority Trial
Ji-Jung JUNG ; Jong-Ho CHEUN ; Soo-Yeon KIM ; Jiwon KOH ; Jai Min RYU ; Tae-Kyung YOO ; Hee-Chul SHIN ; Sung Gwe AHN ; Seho PARK ; Woosung LIM ; Sang-Eun NAM ; Min Ho PARK ; Ku Sang KIM ; Taewoo KANG ; Jeeyeon LEE ; Hyun Jo YOUN ; Yoo Seok KIM ; Chang Ik YOON ; Hong-Kyu KIM ; Hyeong-Gon MOON ; Wonshik HAN ; Nariya CHO ; Min Kyoon KIM ; Han-Byoel LEE
Journal of Breast Cancer 2024;27(1):61-71
Purpose:
Advances in chemotherapeutic and targeted agents have increased pathologic complete response (pCR) rates after neoadjuvant systemic therapy (NST). Vacuum-assisted biopsy (VAB) has been suggested to accurately evaluate pCR. This study aims to confirm the non-inferiority of the 5-year disease-free survival of patients who omitted breast surgery when predicted to have a pCR based on breast magnetic resonance imaging (MRI) and VAB after NST, compared with patients with a pCR who had undergone breast surgery in previous studies.
Methods
The Omission of breast surgery for PredicTed pCR patients wIth MRI and vacuumassisted bIopsy in breaST cancer after neoadjuvant systemic therapy (OPTIMIST) trial is a prospective, multicenter, single-arm, non-inferiority study enrolling in 17 tertiary care hospitals in the Republic of Korea. Eligible patients must have a clip marker placed in the tumor and meet the MRI criteria suggesting complete clinical response (post-NST MRI size ≤ 1 cm and lesion-to-background signal enhancement ratio ≤ 1.6) after NST. Patients will undergo VAB, and breast surgery will be omitted for those with no residual tumor. Axillary surgery can also be omitted if the patient was clinically node-negative before and after NST and met the stringent criteria of MRI size ≤ 0.5 cm. Survival and efficacy outcomes are evaluated over five years.Discussion: This study seeks to establish evidence for the safe omission of breast surgery in exceptional responders to NST while minimizing patient burden. The trial will address concerns about potential undertreatment due to false-negative results and recurrence as well as improved patient-reported quality of life issues from the omission of surgery. Successful completion of this trial may reshape clinical practice for certain breast cancer subtypes and lead to a safe and less invasive approach for selected patients.
5.Masticatory Function, Sex, and Risk of Dementia Among Older Adults:A Population-Based Cohort Study
Dae Jong OH ; Ji Won HAN ; Jun Sung KIM ; Tae Hui KIM ; Kyung Phil KWAK ; Bong Jo KIM ; Shin Gyeom KIM ; Jeong Lan KIM ; Seok Woo MOON ; Joon Hyuk PARK ; Seung-Ho RYU ; Jong Chul YOUN ; Dong Young LEE ; Dong Woo LEE ; Seok Bum LEE ; Jung Jae LEE ; Jin Hyeong JHOO ; Ki Woong KIM
Journal of Korean Medical Science 2024;39(36):e246-
Background:
A decline in masticatory function may indicate brain dysfunction related to dementia, but the relationship between masticatory function and dementia risk remains unclear. This study aimed to investigate whether masticatory function is associated with the risk of cognitive decline and dementia.
Methods:
Data were obtained from the nationwide prospective cohort study of randomly sampled community-dwelling Koreans aged ≥ 60 years. The 5,064 non-demented participants, whose number of chewing cycles per bite was assessed by clinical interview, were followed for 8 years with biennial assessments of cognitive performance and clinical diagnoses of all-cause dementia and Alzheimer’s disease (AD). Structural brain magnetic resonance imaging was collected from a subset of cohort participants and their spouses for imaging analyses.
Results:
Males who chewed ≥ 30 cycles/bite had faster decline in global cognition and memory function and were at higher risk for incident all-cause dementia (hazard ratio [HR], 2.91; 95% confidence interval [CI], 1.18–7.18) and AD (HR, 3.22; 95% CI, 1.14–9.11) compared to males with less than 10 cycles/bite. Additionally, increased chewing cycles in males were associated with reduced brain volume, particularly in regions involved in compensatory cognitive control of mastication. There was no significant association between chewing cycles and the risk of dementia or brain volume in females.
Conclusion
Older men who frequently chew their meals could be considered a notable population at risk for dementia who should be carefully assessed for their cognitive trajectories.
6.Clinical Trial: Efficacy of Mosapride Controlledrelease and Nortriptyline in Patients With Functional Dyspepsia: A Multicenter, Double-placebo, Double-blinded, Randomized Controlled, Parallel Clinical Study
Chung Hyun TAE ; Ra Ri CHA ; Jung-Hwan OH ; Tae-Guen GWEON ; Jong Kyu PARK ; Ki Bae BANG ; Kyung Ho SONG ; Cheal Wung HUH ; Ju Yup LEE ; Cheol Min SHIN ; Jong Wook KIM ; Young Hoon YOUN ; Joong Goo KWON ;
Journal of Neurogastroenterology and Motility 2024;30(1):106-115
Background/Aims:
Prokinetic agents and neuromodulators are among the treatment options for functional dyspepsia (FD), but their comparative efficacy is unclear. We aimed to compare the efficacy of mosapride controlled-release (CR) and nortriptyline in patients with FD after 4 weeks of treatment.
Methods:
Participants with FD were randomly assigned (1:1) to receive mosapride CR (mosapride CR 15 mg and nortriptyline placebo) or nortriptyline (mosapride CR placebo and nortriptyline 10 mg) in double-placebo, double-blinded, randomized controlled, parallel clinical study. The primary endpoint was defined as the proportion of patients with overall dyspepsia improvement after 4 weeks treatment. The secondary endpoints were changes in individual symptom scores, anxiety, depression, and quality of life.
Results:
One hundred nine participants were recruited and assessed for eligibility, and 54 in the mosapride CR group and 50 in the nortriptyline group were included in the modified intention-to-treat protocol. The rate of overall dyspepsia improvement was similar between groups (53.7% vs 54.0%, P = 0.976). There was no difference in the efficacy of mosapride CR and nortriptyline in a subgroup analysis by FD subtype (59.3% vs 52.5% in postprandial distress syndrome, P = 0.615; 44.4% vs 40.0% in epigastric pain syndrome, P = > 0.999; 50.0% vs 59.1% in overlap, P = 0.565; respectively). Both treatments significantly improved anxiety, depression, and quality of life from baseline.
Conclusion
Mosapride CR and nortriptyline showed similar efficacy in patients with FD regardless of the subtype. Both treatments could be equally helpful for improving quality of life and psychological well-being while also relieving dyspepsia.
7.Efficacy of Tegoprazan in Patients With Functional Dyspepsia: A Prospective, Multicenter, Single-arm Study
Cheal Wung HUH ; Young Hoon YOUN ; Da Hyun JUNG ; Ra Ri CHA ; Yeon Ji KIM ; Kyoungwon JUNG ; Kyung Ho SONG ; Ki Bae BANG ; Chung Hyun TAE ; Soo In CHOI ; Cheol Min SHIN ;
Journal of Neurogastroenterology and Motility 2024;30(3):313-321
Background/Aims:
Acid-suppressive drugs, such as proton pump inhibitors (PPIs), are treatment options for functional dyspepsia (FD). However, the efficacy of potassium-competitive acid blockers (P-CABs) in treating FD has not yet been established. This prospective multicenter clinical trial-based study aimed to assess the efficacy and safety of tegoprazan as a P-CAB treatment in patients with FD.
Methods:
FD was diagnosed using the Rome IV criteria. All patients received tegoprazan 50 mg once daily for 8 weeks. Dyspeptic symptoms were assessed using a dyspepsia symptom questionnaire (5-point Likert scale, Nepean Dyspepsia Index-Korean [NDI-K], and gastroesophageal reflux disease–health-related quality of life [GERD-HRQL]). The main outcome was satisfactory symptom relief rates at 8 weeks.
Results:
In this study, from the initial screening of 209 patients, 173 were included in the per-protocol set analysis. Satisfactory symptom relief rates at 8 and 4 weeks were 86.7% and 74.6%, respectively. In addition, the NDI-K and GERD-HRQL scores significantly improved at 8 and 4 weeks compared with the baseline scores. The efficacy of tegoprazan was not influenced by the FD subtype or Helicobacter pylori status. In patients with overlapping FD and GERD, there was a greater improvement in the NDI-K and GERD-HRQL scores than in patients with FD symptoms only. No serious drug-related adverse events occurred during this study.
Conclusion
Tegoprazan (50 mg) administered once daily provided satisfactory symptom relief for FD.
8.Unenhanced Breast MRI With Diffusion-Weighted Imaging for Breast Cancer Detection: Effects of Training on Performance and Agreement of Subspecialty Radiologists
Yeon Soo KIM ; Su Hyun LEE ; Soo-Yeon KIM ; Eun Sil KIM ; Ah Reum PARK ; Jung Min CHANG ; Vivian Youngjean PARK ; Jung Hyun YOON ; Bong Joo KANG ; Bo La YUN ; Tae Hee KIM ; Eun Sook KO ; A Jung CHU ; Jin You KIM ; Inyoung YOUN ; Eun Young CHAE ; Woo Jung CHOI ; Hee Jeong KIM ; Soo Hee KANG ; Su Min HA ; Woo Kyung MOON
Korean Journal of Radiology 2024;25(1):11-23
Objective:
To investigate whether reader training improves the performance and agreement of radiologists in interpreting unenhanced breast magnetic resonance imaging (MRI) scans using diffusion-weighted imaging (DWI).
Materials and Methods:
A study of 96 breasts (35 cancers, 24 benign, and 37 negative) in 48 asymptomatic women was performed between June 2019 and October 2020. High-resolution DWI with b-values of 0, 800, and 1200 sec/mm 2 was performed using a 3.0-T system. Sixteen breast radiologists independently reviewed the DWI, apparent diffusion coefficient maps, and T1-weighted MRI scans and recorded the Breast Imaging Reporting and Data System (BI-RADS) category for each breast. After a 2-h training session and a 5-month washout period, they re-evaluated the BI-RADS categories. A BI-RADS category of 4 (lesions with at least two suspicious criteria) or 5 (more than two suspicious criteria) was considered positive.The per-breast diagnostic performance of each reader was compared between the first and second reviews. Inter-reader agreement was evaluated using a multi-rater κ analysis and intraclass correlation coefficient (ICC).
Results:
Before training, the mean sensitivity, specificity, and accuracy of the 16 readers were 70.7% (95% confidence interval [CI]: 59.4–79.9), 90.8% (95% CI: 85.6–94.2), and 83.5% (95% CI: 78.6–87.4), respectively. After training, significant improvements in specificity (95.2%; 95% CI: 90.8–97.5; P = 0.001) and accuracy (85.9%; 95% CI: 80.9–89.8; P = 0.01) were observed, but no difference in sensitivity (69.8%; 95% CI: 58.1–79.4; P = 0.58) was observed. Regarding inter-reader agreement, the κ values were 0.57 (95% CI: 0.52–0.63) before training and 0.68 (95% CI: 0.62–0.74) after training, with a difference of 0.11 (95% CI: 0.02–0.18; P = 0.01). The ICC was 0.73 (95% CI: 0.69–0.74) before training and 0.79 (95% CI: 0.76–0.80) after training (P = 0.002).
Conclusion
Brief reader training improved the performance and agreement of interpretations by breast radiologists using unenhanced MRI with DWI.
9.Efficacy and safety of sofosbuvir–velpatasvir and sofosbuvir–velpatasvir–voxilaprevir for hepatitis C in Korea: a Phase 3b study
Jeong HEO ; Yoon Jun KIM ; Sung Wook LEE ; Youn-Jae LEE ; Ki Tae YOON ; Kwan Soo BYUN ; Yong Jin JUNG ; Won Young TAK ; Sook-Hyang JEONG ; Kyung Min KWON ; Vithika SURI ; Peiwen WU ; Byoung Kuk JANG ; Byung Seok LEE ; Ju-Yeon CHO ; Jeong Won JANG ; Soo Hyun YANG ; Seung Woon PAIK ; Hyung Joon KIM ; Jung Hyun KWON ; Neung Hwa PARK ; Ju Hyun KIM ; In Hee KIM ; Sang Hoon AHN ; Young-Suk LIM
The Korean Journal of Internal Medicine 2023;38(4):504-513
Despite the availability of direct-acting antivirals (DAAs) for chronic hepatitis C virus (HCV) infection in Korea, need remains for pangenotypic regimens that can be used in the presence of hepatic impairment, comorbidities, or prior treatment failure. We investigated the efficacy and safety of sofosbuvir–velpatasvir and sofosbuvir–velpatasvir–voxilaprevir for 12 weeks in HCV-infected Korean adults. Methods: This Phase 3b, multicenter, open-label study included 2 cohorts. In Cohort 1, participants with HCV genotype 1 or 2 and who were treatment-naive or treatment-experienced with interferon-based treatments, received sofosbuvir–velpatasvir 400/100 mg/day. In Cohort 2, HCV genotype 1 infected individuals who previously received an NS5A inhibitor-containing regimen ≥ 4 weeks received sofosbuvir–velpatasvir–voxilaprevir 400/100/100 mg/day. Decompensated cirrhosis was an exclusion criterion. The primary endpoint was SVR12, defined as HCV RNA < 15 IU/mL 12 weeks following treatment. Results: Of 53 participants receiving sofosbuvir–velpatasvir, 52 (98.1%) achieved SVR12. The single participant who did not achieve SVR12 experienced an asymptomatic Grade 3 ASL/ALT elevation on day 15 and discontinued treatment. The event resolved without intervention. All 33 participants (100%) treated with sofosbuvir–velpatasvir–voxilaprevir achieved SVR 12. Overall, sofosbuvir–velpatasvir and sofosbuvir–velpatasvir–voxilaprevir were safe and well tolerated. Three participants (5.6%) in Cohort 1 and 1 participant (3.0%) in Cohort 2 had serious adverse events, but none were considered treatment-related. No deaths or grade 4 laboratory abnormalities were reported. Conclusions: Treatment with sofosbuvir–velpatasvir or sofosbuvir–velpatasvir–voxilaprevir was safe and resulted in high SVR12 rates in Korean HCV patients.
10.Fibrotic Burden in the Liver Differs Across Metabolic Dysfunction-Associated Fatty Liver Disease Subtypes
Tae Seop LIM ; Ho Soo CHUN ; Soon Sun KIM ; Ja Kyung KIM ; Minjong LEE ; Hyo Jung CHO ; Seung Up KIM ; Jae Youn CHEONG
Gut and Liver 2023;17(4):610-619
Background/Aims:
Metabolic dysfunction-associated fatty liver disease (MAFLD) is categorized into three subtypes: overweight/obese (OW), leanormal weight with metabolic abnormalities, and diabetes mellitus (DM). We investigated whether fibrotic burden in liver differs across subtypes of MAFLD patients.
Methods:
This cross-sectional multicenter study was done in cohorts of subjects who underwent a comprehensive medical health checkup between January 2014 and December 2020. A total of 42,651 patients with ultrasound-diagnosed fatty liver were included. Patients were classified as no MAFLD, OW-MAFLD, lean-MAFLD, and DM-MAFLD. Advanced liver fibrosis was defined based on the nonalcoholic fatty liver disease fibrosis score (NFS) or fibrosis-4 (FIB-4) index.
Results:
The mean age of the patients was 50.0 years, and 74.1% were male. The proportion of patients with NFS-defined advanced liver fibrosis was the highest in DM-MAFLD (6.6%), followed by OW-MAFLD (2.0%), lean-MAFLD (1.3%), and no MAFLD (0.2%). The proportion of patients with FIB-4-defined advanced liver fibrosis was the highest in DM-MAFLD (8.6%), followed by lean-MAFLD (3.9%), OW-MAFLD (3.0%), and no MAFLD (2.0%). With the no MAFLD group as reference, the adjusted odds ratios (95% confidence intervals) for NFS-defined advanced liver fibrosis were 4.46 (2.09 to 9.51), 2.81 (1.12 to 6.39), and 9.52 (4.46 to 20.36) in OW-MAFLD, leanMAFLD, and DM-MAFLD, respectively, and the adjusted odds ratios for FIB-4-defined advanced liver fibrosis were 1.03 (0.78 to 1.36), 1.14 (0.82 to 1.57), and 1.97 (1.48 to 2.62) in OW-MAFLD, lean-MAFLD, and DM-MAFLD.
Conclusions
Fibrotic burden in the liver differs across MAFLD subtypes. Optimized surveillance strategies and therapeutic options might be needed for different MAFLD subtypes.

Result Analysis
Print
Save
E-mail