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.Characteristics of High-Risk Groups for Suicide in Korea Before and After the COVID-19 Pandemic: K-COMPASS Cohort Study
Jeong Hun YANG ; Dae Hun KANG ; C. Hyung Keun PARK ; Min Ji KIM ; Sang Jin RHEE ; Min-Hyuk KIM ; Jinhee LEE ; Sang Yeol LEE ; Won Sub KANG ; Seong-Jin CHO ; Shin Gyeom KIM ; Se-Hoon SHIM ; Jung-Joon MOON ; Jieun YOO ; Weon-Young LEE ; Yong Min AHN
Journal of Korean Neuropsychiatric Association 2024;63(4):246-259
Objectives:
This study examined the changes in the characteristics of high-risk suicide groups in South Korea before and after the COVID-19 pandemic using the Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior (K-COMPASS) cohort.
Methods:
The K-COMPASS is a longitudinal cohort study that started in 2015. The participants included suicide attempters and individuals with suicidal ideation from various hospitals and mental health centers in South Korea. This study compared the sociodemographic and psychiatric characteristics of 800 participants from the first cohort (2015–2019) with 511 participants from the second and third cohorts (2019–2024). Data were collected through structured interviews and validated scales.
Results:
The second and third cohort participants were younger, had a higher proportion of females, and exhibited more severe psychiatric symptoms and higher suicidal risk than the first cohort. The prevalence of physical illnesses decreased, while the use of psychiatric medications and the severity of mental health issues increased. In addition, significant sociodemographic changes were observed, such as higher educational levels and urban residency.
Conclusion
Significant shifts in the characteristics of high-risk suicide groups were observed during the COVID-19 pandemic, highlighting the need for targeted mental health interventions focusing on younger individuals and females to prevent suicide in high-risk groups.
3.Characteristics of High-Risk Groups for Suicide in Korea Before and After the COVID-19 Pandemic: K-COMPASS Cohort Study
Jeong Hun YANG ; Dae Hun KANG ; C. Hyung Keun PARK ; Min Ji KIM ; Sang Jin RHEE ; Min-Hyuk KIM ; Jinhee LEE ; Sang Yeol LEE ; Won Sub KANG ; Seong-Jin CHO ; Shin Gyeom KIM ; Se-Hoon SHIM ; Jung-Joon MOON ; Jieun YOO ; Weon-Young LEE ; Yong Min AHN
Journal of Korean Neuropsychiatric Association 2024;63(4):246-259
Objectives:
This study examined the changes in the characteristics of high-risk suicide groups in South Korea before and after the COVID-19 pandemic using the Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior (K-COMPASS) cohort.
Methods:
The K-COMPASS is a longitudinal cohort study that started in 2015. The participants included suicide attempters and individuals with suicidal ideation from various hospitals and mental health centers in South Korea. This study compared the sociodemographic and psychiatric characteristics of 800 participants from the first cohort (2015–2019) with 511 participants from the second and third cohorts (2019–2024). Data were collected through structured interviews and validated scales.
Results:
The second and third cohort participants were younger, had a higher proportion of females, and exhibited more severe psychiatric symptoms and higher suicidal risk than the first cohort. The prevalence of physical illnesses decreased, while the use of psychiatric medications and the severity of mental health issues increased. In addition, significant sociodemographic changes were observed, such as higher educational levels and urban residency.
Conclusion
Significant shifts in the characteristics of high-risk suicide groups were observed during the COVID-19 pandemic, highlighting the need for targeted mental health interventions focusing on younger individuals and females to prevent suicide in high-risk groups.
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.Characteristics of High-Risk Groups for Suicide in Korea Before and After the COVID-19 Pandemic: K-COMPASS Cohort Study
Jeong Hun YANG ; Dae Hun KANG ; C. Hyung Keun PARK ; Min Ji KIM ; Sang Jin RHEE ; Min-Hyuk KIM ; Jinhee LEE ; Sang Yeol LEE ; Won Sub KANG ; Seong-Jin CHO ; Shin Gyeom KIM ; Se-Hoon SHIM ; Jung-Joon MOON ; Jieun YOO ; Weon-Young LEE ; Yong Min AHN
Journal of Korean Neuropsychiatric Association 2024;63(4):246-259
Objectives:
This study examined the changes in the characteristics of high-risk suicide groups in South Korea before and after the COVID-19 pandemic using the Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior (K-COMPASS) cohort.
Methods:
The K-COMPASS is a longitudinal cohort study that started in 2015. The participants included suicide attempters and individuals with suicidal ideation from various hospitals and mental health centers in South Korea. This study compared the sociodemographic and psychiatric characteristics of 800 participants from the first cohort (2015–2019) with 511 participants from the second and third cohorts (2019–2024). Data were collected through structured interviews and validated scales.
Results:
The second and third cohort participants were younger, had a higher proportion of females, and exhibited more severe psychiatric symptoms and higher suicidal risk than the first cohort. The prevalence of physical illnesses decreased, while the use of psychiatric medications and the severity of mental health issues increased. In addition, significant sociodemographic changes were observed, such as higher educational levels and urban residency.
Conclusion
Significant shifts in the characteristics of high-risk suicide groups were observed during the COVID-19 pandemic, highlighting the need for targeted mental health interventions focusing on younger individuals and females to prevent suicide in high-risk groups.
6.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.
7.Characteristics of High-Risk Groups for Suicide in Korea Before and After the COVID-19 Pandemic: K-COMPASS Cohort Study
Jeong Hun YANG ; Dae Hun KANG ; C. Hyung Keun PARK ; Min Ji KIM ; Sang Jin RHEE ; Min-Hyuk KIM ; Jinhee LEE ; Sang Yeol LEE ; Won Sub KANG ; Seong-Jin CHO ; Shin Gyeom KIM ; Se-Hoon SHIM ; Jung-Joon MOON ; Jieun YOO ; Weon-Young LEE ; Yong Min AHN
Journal of Korean Neuropsychiatric Association 2024;63(4):246-259
Objectives:
This study examined the changes in the characteristics of high-risk suicide groups in South Korea before and after the COVID-19 pandemic using the Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior (K-COMPASS) cohort.
Methods:
The K-COMPASS is a longitudinal cohort study that started in 2015. The participants included suicide attempters and individuals with suicidal ideation from various hospitals and mental health centers in South Korea. This study compared the sociodemographic and psychiatric characteristics of 800 participants from the first cohort (2015–2019) with 511 participants from the second and third cohorts (2019–2024). Data were collected through structured interviews and validated scales.
Results:
The second and third cohort participants were younger, had a higher proportion of females, and exhibited more severe psychiatric symptoms and higher suicidal risk than the first cohort. The prevalence of physical illnesses decreased, while the use of psychiatric medications and the severity of mental health issues increased. In addition, significant sociodemographic changes were observed, such as higher educational levels and urban residency.
Conclusion
Significant shifts in the characteristics of high-risk suicide groups were observed during the COVID-19 pandemic, highlighting the need for targeted mental health interventions focusing on younger individuals and females to prevent suicide in high-risk groups.
8.Psychotropic Medications Promote Time-Dependent Reduction of Suicidal Ideation in Mood Disorder: A Prospective Cohort Study
Jeong Hun YANG ; C. Hyung Keun PARK ; Sang Jin RHEE ; Dae Hun KANG ; Min Ji KIM ; Hyun Jeong LEE ; Sang Yeol LEE ; Se-Hoon SHIM ; Jung-Joon MOON ; Seong-Jin CHO ; Shin Gyeom KIM ; Min-Hyuk KIM ; Jinhee LEE ; Won Sub KANG ; Jieun YOO ; Weon-Young LEE ; Yong Min AHN
Journal of Korean Medical Science 2024;39(31):e226-
Background:
Despite a plethora of research on the topic, there is still no solid evidence that pharmacological treatment actually reduces the risk of suicide in patients with mental illness.In this study, we aimed to assess the effect of psychotropic medications on suicidal ideation in patients with major depressive disorder (MDD) and bipolar disorder (BPD) in two age groups: less than 25 years and 25 years and older.
Methods:
We analyzed 312 patients with mood disorders with current suicidal thoughts or recent suicide attempts. We followed the participants from baseline for 6 months and assessed changes in suicidal ideation with Columbia-Suicide Severity Rating Scale (C-SSRS).The effect of psychotropic drug administration on suicidal ideation over time was analyzed using a linear mixed model.
Results:
In patients aged 25 years and older with mood disorders, suicidal ideation was more severe when using psychotropic drugs than when not using them. However, suicidal ideation decreased rapidly over time. The time-dependent reduction in suicidal ideation was accelerated when using antidepressants and sedatives/hypnotics in adult MDD, and when using mood stabilizers in adult BPD. However, this effect was not observed in participants aged less than 25 years.
Conclusion
Adequate psychotropic medication may reduce suicidal ideation in patients with mood disorders aged 25 years and older. Additional research on psychotropic drugs is needed to effectively reduce the risk of suicide among children and adolescents with mood disorders.
9.Development of a Clinical Guideline for Suicide Prevention in Psychiatric Patients Based on the ADAPTE Methodology
Jeong Hun YANG ; Jieun YOO ; Dae Hun KANG ; C. Hyung Keun PARK ; Sang Jin RHEE ; Min Ji KIM ; Sang Yeol LEE ; Se-Hoon SHIM ; Jung-Joon MOON ; Seong-Jin CHO ; Shin Gyeom KIM ; Min-Hyuk KIM ; Jinhee LEE ; Won Sub KANG ; Weon-Young LEE ; Yong Min AHN
Psychiatry Investigation 2024;21(10):1149-1166
Objective:
Suicide is a significant public health issue, with South Korea having the highest suicide rate among Organisation for Economic Cooperation and Development countries. This study aimed to develop clinical guidelines for suicide prevention in psychiatric patients in Korea using the ADAPTE methodology.
Methods:
The development process involved a comprehensive review of literature, expert consultations, and consensus-building using the Nominal Group Technique and Delphi method. The guidelines focus on evidence-based psychiatric treatments, including both pharmacological and non-pharmacological approaches, tailored to the Korean context. Key findings underscoring the need for standardized treatment protocols for patients with major psychiatric disorders, including bipolar disorder, major depressive disorder, and schizophrenia.
Results:
The guidelines incorporate treatments like lithium, clozapine, atypical antipsychotics, electroconvulsive therapy, and cognitive behavioral therapy, which have shown effectiveness in suicide prevention. Applicability and acceptability within Korea’s healthcare system were addressed, ensuring feasibility given the country’s medical insurance coverage and accessibility. The guidelines were validated through expert reviews and Delphi rounds, achieving consensus on the final recommendations.
Conclusion
The developed guidelines provide a structured, evidence-based approach to reducing suicide rates among psychiatric patients in Korea. Future research will focus on expanding these guidelines to include screening protocols for high-risk groups.
10.Does the dominant hand’s use affect the complication rates in prosthetic breast reconstruction?
Seong Heum JEONG ; Euna HWANG ; Hyun Jeong HA ; Tae Hyun KIM ; Deok-Yeol KIM ; Chung Hun KIM ; Suk Wha KIM ; Hee Chang AHN
Archives of Aesthetic Plastic Surgery 2023;29(4):201-206
Background:
Numerous studies have investigated risk factors for unfavorable outcomes in prosthetic breast reconstruction, such as obesity, perioperative radiotherapy, and acellular dermal matrix use. However, no reports have explored whether the use of the dominant hand influences complications in breast reconstruction. To address this gap in the literature, analyzed complication rates between the dominant and non-dominant sides after reconstruction.
Methods:
We retrospectively reviewed the charts of 160 patients (170 breasts) who underwent breast reconstruction from February 2017 to March 2022. We analyzed the complications between beasts on the dominant and non-dominant sides according to the reconstruction method.
Results:
During prosthetic breast reconstruction, the drainage volume and duration on the dominant side exceeded those on the non-dominant side after reconstruction (duration: 9.79 days on the dominant side vs. 9.12 days on the non-dominant side, P=0.196; volume: 771.1 mL on the dominant side vs. 654.3 mL on the non-dominant side, P=0.027). The incidence of complications such as wound dehiscence, mastectomy flap necrosis, and infection was significantly higher in the dominant hand group (infection: 6 vs. 0, P=0.014; dehiscence: 15 vs. 4, P=0.009; flap necrosis: 13 vs. 4, P=0.024).
Conclusions
Complications including seroma, infection, and mastectomy skin flap necrosis following prosthetic reconstruction were common in breasts on the dominant-hand side. Therefore, meticulous management and restriction of shoulder movement can aid in preventing seroma-related complications in prosthetic breast reconstruction, especially on the side of the dominant hand.

Result Analysis
Print
Save
E-mail