Background: In most industrialized societies, regulations, inspections, insurance, and legal options are established to support workers who suffer injury, disease, or death in relation to their work; in practice, these resources are imperfect or even unavailable due to workplace or employer obstruction. Thus, limitations exist to identify unmet needs in occupational safety and health information. Objective: The aim of this study was to explore hidden issues related to occupational accidents by examining social network services (SNS) data using topic modeling. Methods: Based on the results of a Google search for the phrases occupational accident, industrial accident and occupational diseases, a total of 145 websites were selected. From among these websites, we collected 15,244 documents on queries related to occupational accidents between 2002 and 2018. To transform unstructured text into structure data, natural language processing of the Korean language was conducted. We performed the latent Dirichlet allocation (LDA) as a topic model using a Python library. A time-series linear regression analysis was also conducted to identify yearly trends for the given documents. Results: The results of the LDA model showed 14 topics with 3 themes: workers’ compensation benefits (Theme 1), illicit agreements with the employer (Theme 2), and fatal and non-fatal injuries and vulnerable workers (Theme 3). Theme 1 represented the largest cluster (52.2%) of the collected documents and included keywords related to workers’ compensation (ie, company, occupational injury, insurance, accident, approval, and compensation) and keywords describing specific compensation benefits such as medical expense benefits, temporary incapacity benefits, and disability benefits. In the yearly trend, Theme 1 gradually decreased; however, other themes showed an overall increasing pattern. Certain queries (ie, musculoskeletal system, critical care, and foreign workers) showed no significant variation in the number of queries. Conclusions: We conducted LDA analysis of SNS data of occupational accident–related queries and discovered that the primary concerns of workers posting about occupational injuries and diseases were workers’ compensation benefits, fatal and non-fatal injuries, vulnerable workers, and illicit agreements with employers. While traditional systems focus mainly on quantitative monitoring of occupational accidents, qualitative aspects formulated by topic modeling from unstructured SNS queries may be valuable to address inequalities and improve occupational health and safety.

Background: We hypothesized that analysis of serial ECGs could predict new‐onset atrial fibrillation (AF) more accurately than analysis of a single ECG by detecting the subtle cardiac remodeling that occurs immediately before AF occurrence. Our aim in this study was to compare the performance of 2 types of machine learning (ML) algorithms. Methods and Results: Standard 12‐lead ECGs of patients selected by cardiologists between January 2010 and May 2021 were used for ML model development. Two ML models (single ECG and serial ECG) were developed using a light gradient boosting machine‐learning algorithm. Model performance was evaluated based on the area under the receiver operating characteristic curve, sensitivity, specificity, accuracy, and F1 score. We trained the ML models on 415 964 ECGs from 176 090 patients. When testing the 2 ML models using external validation data sets, the performance of the serial‐ML model was significantly better than that of the single‐ML model for predicting new‐onset AF (single‐ versus serial‐ML model: sensitivity 0.744 versus 0.810; specificity 0.742 versus 0.822; accuracy 0.743 versus 0.816; F1 score 0.743 versus 0.815; area under the receiver operating characteristic curve 0.812 versus 0.880; P <0.001). The Shapley Additive Explanations analysis ranked P‐wave duration and amplitude among the top 10 ECG parameters. Conclusions: An ML model based on serial ECGs from an individual had greater ability to predict new‐onset AF than the ML model based on a single ECG. P‐wave morphologies were associated with future AF prediction.

Background: Although injured employees are legally covered by workers’ compensation insurance in South Korea, some employers make agreements to prevent the injured employees from claiming their compensation. Thus, this leads to underreporting of occupational injury statistics. Illegal compensation (called gong-sang in Korean) is a critical method used to underreport or cover-up occupational injuries. However, gong-sang is not counted in the official occupational injury statistics; therefore, we cannot identify gong-sang–related issues. Objective: This study aimed to analyze social media data using topic modeling to explore hidden knowledge about illegal compensation—gong-sang—for occupational injury in South Korea. Methods: We collected 2210 documents from social media data by filtering the keyword, gong-sang. The study period was between January 1, 2006, and December 31, 2017. After completing natural language processing of the Korean language, a morphological analyzer, we performed topic modeling using latent Dirichlet allocation (LDA) in the Python library, Gensim. A 10-topic model was selected and run with 3000 Gibbs sampling iterations to fit the model. Results: The LDA model was used to classify gong-sang–related documents into 4 categories from a total of 10 topics. Topic 1 was the greatest concern (60.5%). Workers who suffered from industrial accidents seemed to be worried about illegal compensation and legal insurance claims, wherein keywords on the choice between illegal compensation and legal insurance claims were included. In topic 2, keywords were associated with claims for industrial accident insurance benefits. Topics 3 and 4, as the second highest concern (19%), contained keywords implying the monetary compensation of gong-sang. Topics 5 to 10 included keywords on vulnerable jobs (ie, workers in the construction and defense industry, delivery riders, and foreign workers) and body parts (ie, injuries to the hands, face, teeth, lower limbs, and back) to gong-sang. Conclusions: We explored hidden knowledge to identify the salient issues surrounding gong-sang using the LDA model. These topics may provide valuable information to ensure the more efficient operation of South Korea’s occupational health and safety administration and protect vulnerable workers from illegal gong-sang compensation practices.

Background/Objectives: Chronological age (CA) is commonly used in clinical decision-making, yet it may not accurately reflect biological aging. Recent advances in artificial intelligence (AI) allow estimation of electrocardiogram (ECG)-derived heart age, which may serve as a non-invasive biomarker for physiological aging. This study aimed to develop and validate a deep learning model to predict ECG-heart age in individuals with no structural heart disease. Methods: We trained a convolutional neural network (DenseNet-121) using 12-lead ECGs from 292,484 individuals (mean age: 51.4 ± 13.8 years; 42.3% male) without significant cardiac disease. Exclusion criteria included missing age data, age 90 years, and structural abnormalities. CA was used as the target variable. Model performance was evaluated using the coefficient of determination (R²), Pearson correlation coefficient (PCC), mean absolute error (MAE), and root mean square error (RMSE). External validation was conducted using 1191 independent ECGs. Results: The model demonstrated strong predictive performance (R² = 0.783, PCC = 0.885, MAE = 5.023 years, RMSE = 6.389 years). ECG-heart age tended to be overestimated in younger adults (≤30 years) and underestimated in older adults (≥70 years). External validation showed consistent performance (R² = 0.703, PCC = 0.846, MAE = 5.582 years, RMSE = 7.316 years). Conclusions: The proposed AI-based model accurately estimates ECG-heart age in individuals with structurally normal hearts. ECG-derived heart age may serve as a reliable biomarker of biological aging and support future risk stratification strategies.

Kim TM, et al. J Korean Med Sci. 2024 Nov;40:e26. https://doi.org/10.3346/jkms.2025.40.e26

Background: Accurate reports of occupational injuries are important to monitor workplace safety and health initiatives. In South Korea, media reports, experts, and workers have been constantly raising the issue of underreporting. Supposedly it is because employers have strong market “incentives” by underreporting their employees’ injuries. A critical way to underreport or cover-up is illegal compensation (in Korean called “gong-sang”). Unfortunately, “gong-sang” is not counted as official occupational injury statistics. The aim of this study was to analyze the social media data using topic modeling and to explore issues surrounding “gong-sang”. Methods: We used web scraping technology and collected 2,210 social media data from Web search engines. Data was processed to transform unstructured textual documents into structured data using the Python and applied Latent Dirichlet allocation (LDA) in the Python library, Gensim, for topic modeling. Results: Based on the LDA method from “gong-sang”- related documentation, 10 topics were identified. Topic 1 was the greatest concern (60.5%), with keywords implying the choice between illegal compensation (“gong-sang”) and legal insurance claims. The next concern was Topic 2 including keywords associated with claims for industrial accident insurance benefits. The rest topics (topic 3-10) showed the monetary issue, precarious employment, and vulnerable body parts to “gong-sang”. Conclusions: We explored web-based data and identified the salient issues surrounding “gong-sang”. LDA topics may be helpful to ensure efficient occupational health and safety scheme to protect vulnerable employees from “gong-sang” practices. Key messages: The topics formulated by LDA included queries about legal insurance claims. Legal insurance claims including private or social insurance, monetary compensation, injured body parts, and the type of jobs vulnerable to “gong-sang”.