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.Discordance Between Spine-Hip and Paretic-Nonparetic Hip Bone Mineral Density in Hemiplegic Stroke Patients: A Multicenter Retrospective Study
Seung Don YOO ; Tae-Woo KIM ; Byung-Mo OH ; Seung Ah LEE ; Chanwoo KIM ; Ho Yeon CHUNG ; Jung Eun SON ; Ji Yeon LEE ; Hyunji LEE ; Hoo Young LEE
Annals of Rehabilitation Medicine 2024;48(6):413-422
Objective:
To identify the prevalence and factors associated with T-score discordance between the spine and hip, as well as between the paretic and non-paretic hips in hemiplegic stroke patients, this study investigated bone mineral density (BMD) patterns. Bone loss predominantly affects the paretic hip after a stroke, and typical clinical assessments using dual-energy X-ray absorptiometry (DXA) that scan the lumbar spine (LS) and a single hip may overlook an osteoporosis diagnosis. This oversight could potentially lead to suboptimal treatment for stroke patients.
Methods:
This study was a multicenter retrospective analysis of 540 patients admitted for stroke rehabilitation between October 2014 and February 2022, who underwent DXA of LS and bilateral hips.
Results:
The prevalence rates of concordance, low LS discordance, and low hip discordance between the LS and hips were 48.2%, 12.2%, and 39.6%, respectively. The discordance rate between bilateral hips was 17.0%. The paretic side had significantly lower total hip T-scores than the non-paretic side (p<0.001). Notably low paretic hip discordance was more prevalent during the chronic phase. DXA scans of the LS and both hips revealed a 0.7%–0.9% higher major discordance compared to LS and single hip DXA scans. The multivariate analysis revealed a significant correlation between a low paretic hip discordance and cognitive impairment (adjusted odds ratio 0.071, 95% confidence interval 0.931–1.003, p<0.05).
Conclusion
Since stroke survivors are at high risk for hip fractures, comprehensive BMD assessments, which include LS and bilateral hips, should be considered for post-stroke osteoporosis care to enhance diagnostic accuracy and timely treatment.
3.Dental Age Estimation in Children Using Convolution Neural Network Algorithm: A Pilot Study
Byung-Yoon ROH ; Hyun-Jeong PARK ; Kyung-Ryoul KIM ; In-Soo SEO ; Yeon-Ho OH ; Ju-Heon LEE ; Chang-Un CHOI ; Yo-Seob SEO ; Ji-Won RYU ; Jong-Mo AHN
Journal of Oral Medicine and Pain 2024;49(4):118-123
Purpose:
Recently, deep learning techniques have been introduced for age estimation, with automated methods based on radiographic analysis demonstrating high accuracy. In this study, we applied convolutional neural network (CNN) techniques to the lower dentition area on orthopantomograms (OPGs) of children to develop an automated age estimation model and evaluate its accuracy for use in forensic dentistry.
Methods:
In this study, OPGs of 2,856 subjects aged 3-14 years were analyzed. The You Only Look Once (YOLO) V8 object detection technique was applied to extract the mandibular dentition area on OPGs, designating it as the region of interest (ROI). First, 200 radiographs were randomly selected, and were used to train a model for extracting the ROI. The trained model was then applied to the entire dataset. For the CNN image classification task, 80% of OPGs were allocated to the training set, while the remaining 20% were used as the test set. A transfer learning approach was employed using the ResNet50 and VGG19 backbone models, with an ensemble technique combining these models to improve performance. The mean absolute error (MAE) on the test set was used as the validation metric, and the model with the lowest MAE was selected.
Results:
In this study, the age estimation model developed using mandibular dentition region from OPGs achieved MAE and root mean squared error (RMSE) values of 0.501 and 0.742, respectively, on the test set, and MAE and RMSE values of 0.273 and 0.354, respectively, on the training set.
Conclusions
The automated age estimation model developed in this study demonstrated accuracy comparable to that of previous research and shows potential for applications in forensic investigations. Increasing the sample size and incorporating diverse deep learning techniques are expected to further enhance the accuracy of future age estimation models.
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.Discordance Between Spine-Hip and Paretic-Nonparetic Hip Bone Mineral Density in Hemiplegic Stroke Patients: A Multicenter Retrospective Study
Seung Don YOO ; Tae-Woo KIM ; Byung-Mo OH ; Seung Ah LEE ; Chanwoo KIM ; Ho Yeon CHUNG ; Jung Eun SON ; Ji Yeon LEE ; Hyunji LEE ; Hoo Young LEE
Annals of Rehabilitation Medicine 2024;48(6):413-422
Objective:
To identify the prevalence and factors associated with T-score discordance between the spine and hip, as well as between the paretic and non-paretic hips in hemiplegic stroke patients, this study investigated bone mineral density (BMD) patterns. Bone loss predominantly affects the paretic hip after a stroke, and typical clinical assessments using dual-energy X-ray absorptiometry (DXA) that scan the lumbar spine (LS) and a single hip may overlook an osteoporosis diagnosis. This oversight could potentially lead to suboptimal treatment for stroke patients.
Methods:
This study was a multicenter retrospective analysis of 540 patients admitted for stroke rehabilitation between October 2014 and February 2022, who underwent DXA of LS and bilateral hips.
Results:
The prevalence rates of concordance, low LS discordance, and low hip discordance between the LS and hips were 48.2%, 12.2%, and 39.6%, respectively. The discordance rate between bilateral hips was 17.0%. The paretic side had significantly lower total hip T-scores than the non-paretic side (p<0.001). Notably low paretic hip discordance was more prevalent during the chronic phase. DXA scans of the LS and both hips revealed a 0.7%–0.9% higher major discordance compared to LS and single hip DXA scans. The multivariate analysis revealed a significant correlation between a low paretic hip discordance and cognitive impairment (adjusted odds ratio 0.071, 95% confidence interval 0.931–1.003, p<0.05).
Conclusion
Since stroke survivors are at high risk for hip fractures, comprehensive BMD assessments, which include LS and bilateral hips, should be considered for post-stroke osteoporosis care to enhance diagnostic accuracy and timely treatment.
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.Discordance Between Spine-Hip and Paretic-Nonparetic Hip Bone Mineral Density in Hemiplegic Stroke Patients: A Multicenter Retrospective Study
Seung Don YOO ; Tae-Woo KIM ; Byung-Mo OH ; Seung Ah LEE ; Chanwoo KIM ; Ho Yeon CHUNG ; Jung Eun SON ; Ji Yeon LEE ; Hyunji LEE ; Hoo Young LEE
Annals of Rehabilitation Medicine 2024;48(6):413-422
Objective:
To identify the prevalence and factors associated with T-score discordance between the spine and hip, as well as between the paretic and non-paretic hips in hemiplegic stroke patients, this study investigated bone mineral density (BMD) patterns. Bone loss predominantly affects the paretic hip after a stroke, and typical clinical assessments using dual-energy X-ray absorptiometry (DXA) that scan the lumbar spine (LS) and a single hip may overlook an osteoporosis diagnosis. This oversight could potentially lead to suboptimal treatment for stroke patients.
Methods:
This study was a multicenter retrospective analysis of 540 patients admitted for stroke rehabilitation between October 2014 and February 2022, who underwent DXA of LS and bilateral hips.
Results:
The prevalence rates of concordance, low LS discordance, and low hip discordance between the LS and hips were 48.2%, 12.2%, and 39.6%, respectively. The discordance rate between bilateral hips was 17.0%. The paretic side had significantly lower total hip T-scores than the non-paretic side (p<0.001). Notably low paretic hip discordance was more prevalent during the chronic phase. DXA scans of the LS and both hips revealed a 0.7%–0.9% higher major discordance compared to LS and single hip DXA scans. The multivariate analysis revealed a significant correlation between a low paretic hip discordance and cognitive impairment (adjusted odds ratio 0.071, 95% confidence interval 0.931–1.003, p<0.05).
Conclusion
Since stroke survivors are at high risk for hip fractures, comprehensive BMD assessments, which include LS and bilateral hips, should be considered for post-stroke osteoporosis care to enhance diagnostic accuracy and timely treatment.
8.Practice guidelines for managing extrahepatic biliary tract cancers
Hyung Sun KIM ; Mee Joo KANG ; Jingu KANG ; Kyubo KIM ; Bohyun KIM ; Seong-Hun KIM ; Soo Jin KIM ; Yong-Il KIM ; Joo Young KIM ; Jin Sil KIM ; Haeryoung KIM ; Hyo Jung KIM ; Ji Hae NAHM ; Won Suk PARK ; Eunkyu PARK ; Joo Kyung PARK ; Jin Myung PARK ; Byeong Jun SONG ; Yong Chan SHIN ; Keun Soo AHN ; Sang Myung WOO ; Jeong Il YU ; Changhoon YOO ; Kyoungbun LEE ; Dong Ho LEE ; Myung Ah LEE ; Seung Eun LEE ; Ik Jae LEE ; Huisong LEE ; Jung Ho IM ; Kee-Taek JANG ; Hye Young JANG ; Sun-Young JUN ; Hong Jae CHON ; Min Kyu JUNG ; Yong Eun CHUNG ; Jae Uk CHONG ; Eunae CHO ; Eui Kyu CHIE ; Sae Byeol CHOI ; Seo-Yeon CHOI ; Seong Ji CHOI ; Joon Young CHOI ; Hye-Jeong CHOI ; Seung-Mo HONG ; Ji Hyung HONG ; Tae Ho HONG ; Shin Hye HWANG ; In Gyu HWANG ; Joon Seong PARK
Annals of Hepato-Biliary-Pancreatic Surgery 2024;28(2):161-202
Background:
s/Aims: Reported incidence of extrahepatic bile duct cancer is higher in Asians than in Western populations. Korea, in particular, is one of the countries with the highest incidence rates of extrahepatic bile duct cancer in the world. Although research and innovative therapeutic modalities for extrahepatic bile duct cancer are emerging, clinical guidelines are currently unavailable in Korea. The Korean Society of Hepato-Biliary-Pancreatic Surgery in collaboration with related societies (Korean Pancreatic and Biliary Surgery Society, Korean Society of Abdominal Radiology, Korean Society of Medical Oncology, Korean Society of Radiation Oncology, Korean Society of Pathologists, and Korean Society of Nuclear Medicine) decided to establish clinical guideline for extrahepatic bile duct cancer in June 2021.
Methods:
Contents of the guidelines were developed through subgroup meetings for each key question and a preliminary draft was finalized through a Clinical Guidelines Committee workshop.
Results:
In November 2021, the finalized draft was presented for public scrutiny during a formal hearing.
Conclusions
The extrahepatic guideline committee believed that this guideline could be helpful in the treatment of patients.
9.Effect of calcium silicate-based sealer to bone tissue of mandible of rats
Jee-Seon TAE ; Ki-Yeon YOO ; Jin-Woo KIM ; Kyung-Mo CHO ; Yoon LEE ; Se-Hee PARK
Journal of Dental Rehabilitation and Applied Science 2024;40(1):1-12
Purpose:
To histologically evaluate the effects of three calcium silicate-based sealers on rat mandible tissue.
Materials and Methods:
Rats were randomly divided as follows: A group that sacrificed immediately after cavity preparation, a group that sacri-ficed two weeks after cavity preparation, a group that sacrificed two weeks after CeraSeal (CS), AH Plus Bioceramic (AHB), or OneFil (OF) sealer injection, respectively. After tissue processing for all groups, the bone tissue area (%) and the number of osteoclasts in and around the cavity were measured under a microscope. The results of each group were compared and statistical analysis was performed using one-way ANOVA and Tukey’s test.
Results:
The formation of bone tissue and the presence of osteoclasts in the cav-ity were observed in the group that sacrificed two weeks after cavity preparation and the group sacrificed two weeks after AHB seal-er injection, and these groups showed significantly higher average bone tissue area (%) than the other groups. In the other groups, no inflammation or foreign body reaction occurred in the cavity, and no osteoclasts were observed.
Conclusion
All calcium silicatebased sealers used in this study showed a favorable bone tissue response when injected into the rat mandible. In particular, higher bone formation in the cavity was observed in AHB.
10.Clinical Characteristics and Long-term Outcomes of Pediatric Ulcerative Colitis: A Single-Center Experience in Korea
Jooyoung JANG ; Sung Hee LEE ; In Sook JEONG ; Jinmin CHO ; Hyun Jin KIM ; Seak Hee OH ; Dae Yeon KIM ; Ho-Su LEE ; Sang Hyoung PARK ; Byong Duk YE ; Suk-Kyun YANG ; Kyung Mo KIM
Gut and Liver 2022;16(2):236-245
Background/Aims:
Although pediatric ulcerative colitis (UC) has a different phenotype and clinical course than adult UC, its clinical features and outcomes are poorly defined, especially in Asian populations. This study investigated the clinical features and long-term outcomes of pediatric UC in a Korean population.
Methods:
We retrospectively analyzed 208 patients aged <18 years diagnosed with UC between 1987 and 2013. The patient characteristics at diagnosis according to the Paris classification and the clinical course were analyzed.
Results:
The male-to-female ratio was 1.3:1, and the median patient age was 15.5 years. At diagnosis, 28.8% of patients had proctitis (E1), 27.8%, left-sided colitis (E2); 5.2%, extensive colitis (E3); and 38.2%, pancolitis (E4). The cumulative probabilities of extension after 5, 10, 15, and 20 years were 32.7%, 40.4%, 52.5%, and 65.8%, respectively. Eighteen patients underwent colectomy, and three patients had colorectal cancer. The cumulative probabilities of colectomy after 5, 10, 15, and 20 years were 7.1%, 8.9%, 12.6%, and 15.6%, and those of colorectal cancer after 10, 15, and 20 years were 0%, 2.1%, and 12.0%, respectively. The disease extent, Pediatric Ulcerative Colitis Activity Index severity, and systemic corticosteroid therapy were significant risk factors for colectomy. The development of primary sclerosing cholangitis was significantly associated with colorectal cancer.
Conclusions
This study provides detailed information on the disease phenotype and long-term clinical outcomes in a large cohort of Korean children with UC. They have extensive disease at diagnosis, a high rate of disease extension, and a low rate of cumulative colectomy.

Result Analysis
Print
Save
E-mail