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.Oncological Outcomes in Men with Metastatic Castration-Resistant Prostate Cancer Treated with Enzalutamide with versus without Confirmatory Bone Scan
Chang Wook JEONG ; Jang Hee HAN ; Dong Deuk KWON ; Jae Young JOUNG ; Choung-Soo KIM ; Hanjong AHN ; Jun Hyuk HONG ; Tae-Hwan KIM ; Byung Ha CHUNG ; Seong Soo JEON ; Minyong KANG ; Sung Kyu HONG ; Tae Young JUNG ; Sung Woo PARK ; Seok Joong YUN ; Ji Yeol LEE ; Seung Hwan LEE ; Seok Ho KANG ; Cheol KWAK
Cancer Research and Treatment 2024;56(2):634-641
Purpose:
In men with metastatic castration-resistant prostate cancer (mCRPC), new bone lesions are sometimes not properly categorized through a confirmatory bone scan, and clinical significance of the test itself remains unclear. This study aimed to demonstrate the performance rate of confirmatory bone scans in a real-world setting and their prognostic impact in enzalutamide-treated mCRPC.
Materials and Methods:
Patients who received oral enzalutamide for mCRPC during 2014-2017 at 14 tertiary centers in Korea were included. Patients lacking imaging assessment data or insufficient drug exposure were excluded. The primary outcome was overall survival (OS). Secondary outcomes included performance rate of confirmatory bone scans in a real-world setting. Kaplan-Meier analysis and multivariate Cox regression analysis were performed.
Results:
Overall, 520 patients with mCRPC were enrolled (240 [26.2%] chemotherapy-naïve and 280 [53.2%] after chemotherapy). Among 352 responders, 92 patients (26.1%) showed new bone lesions in their early bone scan. Confirmatory bone scan was performed in 41 patients (44.6%), and it was associated with prolonged OS in the entire population (median, 30.9 vs. 19.7 months; p < 0.001), as well as in the chemotherapy-naïve (median, 47.2 vs. 20.5 months; p=0.011) and post-chemotherapy sub-groups (median, 25.5 vs. 18.0 months; p=0.006). Multivariate Cox regression showed that confirmatory bone scan performance was an independent prognostic factor for OS (hazard ratio 0.35, 95% confidence interval, 0.18 to 0.69; p=0.002).
Conclusion
Confirmatory bone scan performance was associated with prolonged OS. Thus, the premature discontinuation of enzalutamide without confirmatory bone scans should be discouraged.
5.Stroke-Specific Predictors of Major Bleeding in Anticoagulated Patients With Stroke and Atrial Fibrillation: A Nationwide Multicenter Registry-Based Study
Darda CHUNG ; Tae-Jin SONG ; Bum Joon KIM ; Sung Hyuk HEO ; Jin-Man JUNG ; Kyungmi OH ; Chi Kyung KIM ; Sungwook YU ; Kwang Yeol PARK ; Jeong-Min KIM ; Jong-Ho PARK ; Man-Seok PARK ; Joon-Tae KIM ; Yang-Ha HWANG ; Yong-Jae KIM ; Jong-Won CHUNG ; Oh Young BANG ; Gyeong-Moon KIM ; Woo-Keun SEO ; Jay Chol CHOI
Journal of Clinical Neurology 2023;19(5):429-437
Background:
and Purpose The congestive heart failure, hypertension, age, diabetes, previous stroke/transient ischemic attack (CHA2DS2-VASc) and hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly, drugs/alcohol (HAS-BLED) scores have been validated in estimating the risks of ischemic stroke and major bleeding, respectively, in patients with atrial fibrillation (AF). This study investigated stroke-specific predictors of major bleeding in patients with stroke and AF who were taking oral anticoagulants (OACs).
Methods:
Subjects were selected from patients enrolled in the Korean ATrial fibrillaTion EvaluatioN regisTry in Ischemic strOke patieNts (K-ATTENTION) nationwide multicenter registry between 2013 and 2015. Patients were excluded if they were not taking OACs, had no brain imaging data, or had intracranial bleeding directly related to the index stroke. Major bleeding was defined according to International Society of Thrombosis and Haemostasis criteria. Cox regression analyses were performed to assess the associations between clinical variables and major bleeding and Kaplan-Meier estimates were performed to analyze event-free survival.
Results:
Of a total of 3,213 patients, 1,414 subjects (mean age of 72.6 years, 52.5% males) were enrolled in this study. Major bleeding was reported in 34 patients during the median follow-up period of 1.73 years. Multivariable analysis demonstrated that initial National Institutes of Health Stroke Scale scores (hazard ratio [HR] 1.07, p=0.006), hypertension (HR 3.18, p=0.030), persistent AF type (HR 2.51, p=0.016), and initial hemoglobin level (HR 0.74, p=0.001) were independently associated with major bleeding risk. Except for hypertension, these associations remained significant after adjusting for the HAS-BLED score. Intracranial atherosclerosis presented a trend of association without statistical significance (HR 2.21, p=0.050).
Conclusions
This study found that major bleeding risk was independently associated with stroke-specific factors in anticoagulated patients with stroke and AF. This has the clinical implication that baseline characteristics of patients with stroke and AF should be considered in secondary prevention, which would bring the net clinical benefit of balancing recurrent stroke prevention with minimal bleeding complications.
6.Therapeutic Effect of Multilevel Surgery on Laryngopharyngeal Reflux in Obstructive Sleep Apnea Patients: Impact on the Reflux Symptom Index and Reflux Finding Score
Bokhyun SONG ; Yong Gi JUNG ; Sang Duk HONG ; Eun Kyu LEE ; Byung Kil KIM ; Song I PARK ; Sung Ha JUNG ; Gwanghui RYU ; Hyo Yeol KIM
Clinical and Experimental Otorhinolaryngology 2022;15(4):346-353
Objectives:
. Our previous study found that multilevel obstructive sleep apnea (OSA) surgery mitigated laryngopharyngeal reflux (LPR) symptoms in terms of the reflux symptom index (RSI), but no studies have investigated the impact of OSA surgery on laryngoscopic parameters. The aim of this study was to examine the clinical outcome of LPR improvement following OSA surgery, with a focus on both the RSI and the reflux finding score (RFS).
Methods:
. Prospectively collected data from 28 patients who underwent multilevel OSA surgery from 2017 to 2021 were retrospectively analyzed. Patients were asked to complete the RSI questionnaire and underwent a laryngoscopic examination to evaluate the RFS before and after surgery. Age, height, weight, body mass index (BMI), and polysomnography data before and after surgery were also reviewed.
Results:
. After surgery, the total RSI and RFS decreased significantly from 11.96±8.40 to 7.68±6.82 (P=0.003) and from 6.57±3.49 to 3.21±1.87 (P<0.001). The positive rates of RSI and RFS decreased from 28.6% to 17.9% and 32.1% to 0%, respectively. Significant improvements were found in the RSI subdomains of throat clearing, throat mucus, breathing difficulty, troublesome cough, and heartburn sensation, while all RFS subdomains except granuloma improved significantly. In subgroup analyses, no significant differences were found between subgroups based on age, OSA severity, or BMI.
Conclusion
. OSA surgery has the potential to alleviate both LPR symptoms and laryngoscopic results. Additional research integrating more objective techniques and novel treatment strategies is required to better comprehend the clinical impact of OSA surgery on LPR.
7.A portrait drawing of the 17th century Korean scholar based on craniofacial reconstruction
Joon Yeol RYU ; A Young YOON ; Yeon Kyung PARK ; Won Joon LEE ; Mi Kyung SONG ; Jong Ha HONG ; Eun Jin PARK ; Soon Chul CHA ; Dongsoo YOO ; Myeung Ju KIM ; Dong Hoon SHIN
Anatomy & Cell Biology 2022;55(4):512-519
As a technique mainly hiring in forensic investigation field to identify the descents, craniofacial reconstruction (CFR) is also used in archaeology to create the faces from ancient or medieval human remains, when there is little information about his/her appearance. Eung-Cheok Ko (1531–1605) was a writer and scholar in the mid Joseon period. In January of 2019, His mummified body was found at Gumi, Kyeonsangbuk-do, Korea. The remains were anthropologically examined, and archaeological CFR was also requested for this case. This report reveals the case’s facial reconstruction process and his portrait that is drawn based on the 3-dimensional CFR result.
8.Lysozyme hydrochloride 0.01%, sodium fluoride 0.02%, cetylpyridinium chloride 0.05% antibacterial and sterilizing effect of mouth freshener
Hye-Won PARK ; Ja-Won CHO ; Hyun-Jun YOO ; Yeol-Mae JEON ; Kyong-Hoon SHIN ; Seong-Min HA
Journal of Korean Academy of Oral Health 2022;46(2):70-77
Objectives:
This study aimed to assess the antibacterial, bactericidal, and mouth freshener effects of lysozyme hydrochloride 0.01%, sodium fluoride 0.02%, and cetylpyridinium chloride 0.05%.
Methods:
Eight oral disease-related bacteria were cultivated anaerobically. Four samples were prepared with or without 0.5% cetylpyridinium chloride, 0.2% sodium fluoride, and 0.1% lysozyme hydrochloride. Antimicrobial activity was tested in 96-well microplates. After assessing the bacterial count, the bacterial suspension was mixed with samples and spread on agar. The bactericidal rate was calculated by counting and comparing treated and untreated colonies.
Results:
Lysozyme hydrochloride 0.01%, sodium fluoride 0.02%, and cetylpyridinium chloride 0.05% mouth fresheners sterilized 99.99% of 8 oral bacteria, including Streprococcus mutans. Lysozyme hydrochloride 0.01%, sodium fluoride 0.02%, and cetylpyridinium chloride 0.05% mouth fresheners showed 99.97% bactericidal activity against Lactobacillus acidophilus.
Conclusions
Lysozyme hydrochloride 0.01%, sodium fluoride 0.02%, and cetylpyridinium chloride 0.05% mouth fresheners confirmed the sterilization and antibacterial effects on oral disease-causing bacteria.
9.Initiation of Guideline-Matched Oral Anticoagulant in Atrial Fibrillation-Related Stroke
Mi-Yeon EUN ; Jae-Young KIM ; Yang-Ha HWANG ; Man-Seok PARK ; Joon-Tae KIM ; Kang-Ho CHOI ; Jin-Man JUNG ; Sungwook YU ; Chi Kyung KIM ; Kyungmi OH ; Tae-Jin SONG ; Yong-Jae KIM ; Bum Joon KIM ; Sung Hyuk HEO ; Kwang-Yeol PARK ; Jeong-Min KIM ; Jong-Ho PARK ; Jay Chol CHOI ; Jong-Won CHUNG ; Oh Young BANG ; Gyeong-Moon KIM ; Woo-Keun SEO
Journal of Stroke 2021;23(1):113-123
Background:
and Purpose To evaluate the outcome events and bleeding complications of the European Society of Cardiology (ESC) guideline-matched oral anticoagulant therapy for patients with acute ischemic stroke and atrial fibrillation (AF).
Methods:
Patients with acute ischemic stroke and AF from a nationwide multicenter registry (Korean ATrial fibrillaTion EvaluatioN regisTry in Ischemic strOke patieNts [K-ATTENTION]) between January 2013 and December 2015 were included in the study. Patients were divided into the ESC guideline-matched and the non-matched groups. The primary outcome was recurrence of any stroke during the 90-day follow-up period. Secondary outcomes were major adverse cerebrovascular and cardiovascular events, ischemic stroke, intracranial hemorrhage, acute coronary syndrome, allcause mortality, and major hemorrhage. Propensity score matching and logistic regression analyses were performed to assess the effect of the treatments administered.
Results:
Among 2,321 eligible patients, 1,126 patients were 1:1 matched to the ESC guidelinematched and the non-matched groups. As compared with the non-matched group, the ESC guideline-matched group had a lower risk of any recurrent stroke (1.4% vs. 3.4%; odds ratio [OR], 0.41; 95% confidence interval [CI], 0.18 to 0.95). The risk of recurrent ischemic stroke was lower in the ESC guideline-matched group than in the non-matched group (0.9% vs. 2.7%; OR, 0.32; 95% CI, 0.11 to 0.88). There was no significant difference in the other secondary outcomes between the two groups.
Conclusions
ESC guideline-matched oral anticoagulant therapy was associated with reduced risks of any stroke and ischemic stroke as compared with the non-matched therapy.
10.A Rare Case of Pituicytoma Presenting Cystic Formation
Young Soo HA ; Soo Eon LEE ; Sung Choon PARK ; Sung Yeol AHN ; Yoon Yang JUNG
Brain Tumor Research and Treatment 2021;9(2):70-74
Pituicytoma is a rare solid benign tumor of the sellar and/or suprasellar region originating from the pituicytes of the neurohypophysis or infundibulum, which is not differentiated from a pituitary adenoma that is diagnosed mostly in the sellar and/or suprasellar region. In addition, cystic tumors are very rare and have not been reported due to their solid and hypervascular natures. A 33-year-old man presented with a chronic headache which exacerbated recently. MRI was performed and revealed a cystic tumor in the sellar and suprasellar regions with a small parenchymal island in the cyst compressing the optic chiasm. The endoscopic endonasal transsphenoidal approach was used to remove the tumor. Immunohistochemical staining was positive for thyroid transcription factor 1, S-100 protein, and glial fibrillary acidic protein. The pituicytoma was diagnosed based on histologic findings. The authors review herein the literature on clinical presentation, diagnosis, surgical management, and outcome.

Result Analysis
Print
Save
E-mail