Purpose: Accurate prediction of blood glucose variability is crucial for effective diabetes management, as both hypoglycemia and hyperglycemia are associated with increased morbidity and mortality. However, conventional predictive models rely primarily on patient-specific biometric data, often neglecting the influence of patient–provider interactions, which can significantly impact outcomes. This study introduces Cyclic Dual Latent Discovery (CDLD), a deep learning framework that explicitly models patient–provider interactions to improve prediction of blood glucose levels. By leveraging a real-world intensive care unit (ICU) dataset, the model captures latent attributes of both patients and providers, thus improving forecasting accuracy. Methods: ICU patient records were obtained from the MIMIC-IV v3.0 critical care database, including approximately 5,014 instances of patient–provider interaction. The CDLD model uses a cyclic training mechanism that alternately updates patient and provider latent representations to optimize predictive performance. During preprocessing, all numeric features were normalized, and extreme glucose values were capped at 500 mg/dL to mitigate the effect of outliers. Results: CDLD outperformed conventional models, achieving a root mean square error of 0.0852 on the validation set and 0.0899 on the test set, which indicates improved generalization. The model effectively captured latent patient–provider interaction patterns, yielding more accurate glucose variability predictions than baseline approaches. Conclusion Integrating patient–provider interaction modeling into predictive frameworks can increase blood glucose prediction accuracy. The CDLD model offers a novel approach to diabetes management, potentially paving the way for artificial intelligence-driven personalized treatment strategies.

Purpose: Accurate prediction of blood glucose variability is crucial for effective diabetes management, as both hypoglycemia and hyperglycemia are associated with increased morbidity and mortality. However, conventional predictive models rely primarily on patient-specific biometric data, often neglecting the influence of patient–provider interactions, which can significantly impact outcomes. This study introduces Cyclic Dual Latent Discovery (CDLD), a deep learning framework that explicitly models patient–provider interactions to improve prediction of blood glucose levels. By leveraging a real-world intensive care unit (ICU) dataset, the model captures latent attributes of both patients and providers, thus improving forecasting accuracy. Methods: ICU patient records were obtained from the MIMIC-IV v3.0 critical care database, including approximately 5,014 instances of patient–provider interaction. The CDLD model uses a cyclic training mechanism that alternately updates patient and provider latent representations to optimize predictive performance. During preprocessing, all numeric features were normalized, and extreme glucose values were capped at 500 mg/dL to mitigate the effect of outliers. Results: CDLD outperformed conventional models, achieving a root mean square error of 0.0852 on the validation set and 0.0899 on the test set, which indicates improved generalization. The model effectively captured latent patient–provider interaction patterns, yielding more accurate glucose variability predictions than baseline approaches. Conclusion Integrating patient–provider interaction modeling into predictive frameworks can increase blood glucose prediction accuracy. The CDLD model offers a novel approach to diabetes management, potentially paving the way for artificial intelligence-driven personalized treatment strategies.

Purpose: Accurate prediction of blood glucose variability is crucial for effective diabetes management, as both hypoglycemia and hyperglycemia are associated with increased morbidity and mortality. However, conventional predictive models rely primarily on patient-specific biometric data, often neglecting the influence of patient–provider interactions, which can significantly impact outcomes. This study introduces Cyclic Dual Latent Discovery (CDLD), a deep learning framework that explicitly models patient–provider interactions to improve prediction of blood glucose levels. By leveraging a real-world intensive care unit (ICU) dataset, the model captures latent attributes of both patients and providers, thus improving forecasting accuracy. Methods: ICU patient records were obtained from the MIMIC-IV v3.0 critical care database, including approximately 5,014 instances of patient–provider interaction. The CDLD model uses a cyclic training mechanism that alternately updates patient and provider latent representations to optimize predictive performance. During preprocessing, all numeric features were normalized, and extreme glucose values were capped at 500 mg/dL to mitigate the effect of outliers. Results: CDLD outperformed conventional models, achieving a root mean square error of 0.0852 on the validation set and 0.0899 on the test set, which indicates improved generalization. The model effectively captured latent patient–provider interaction patterns, yielding more accurate glucose variability predictions than baseline approaches. Conclusion Integrating patient–provider interaction modeling into predictive frameworks can increase blood glucose prediction accuracy. The CDLD model offers a novel approach to diabetes management, potentially paving the way for artificial intelligence-driven personalized treatment strategies.

Purpose: Accurate prediction of blood glucose variability is crucial for effective diabetes management, as both hypoglycemia and hyperglycemia are associated with increased morbidity and mortality. However, conventional predictive models rely primarily on patient-specific biometric data, often neglecting the influence of patient–provider interactions, which can significantly impact outcomes. This study introduces Cyclic Dual Latent Discovery (CDLD), a deep learning framework that explicitly models patient–provider interactions to improve prediction of blood glucose levels. By leveraging a real-world intensive care unit (ICU) dataset, the model captures latent attributes of both patients and providers, thus improving forecasting accuracy. Methods: ICU patient records were obtained from the MIMIC-IV v3.0 critical care database, including approximately 5,014 instances of patient–provider interaction. The CDLD model uses a cyclic training mechanism that alternately updates patient and provider latent representations to optimize predictive performance. During preprocessing, all numeric features were normalized, and extreme glucose values were capped at 500 mg/dL to mitigate the effect of outliers. Results: CDLD outperformed conventional models, achieving a root mean square error of 0.0852 on the validation set and 0.0899 on the test set, which indicates improved generalization. The model effectively captured latent patient–provider interaction patterns, yielding more accurate glucose variability predictions than baseline approaches. Conclusion Integrating patient–provider interaction modeling into predictive frameworks can increase blood glucose prediction accuracy. The CDLD model offers a novel approach to diabetes management, potentially paving the way for artificial intelligence-driven personalized treatment strategies.

Purpose: Accurate prediction of blood glucose variability is crucial for effective diabetes management, as both hypoglycemia and hyperglycemia are associated with increased morbidity and mortality. However, conventional predictive models rely primarily on patient-specific biometric data, often neglecting the influence of patient–provider interactions, which can significantly impact outcomes. This study introduces Cyclic Dual Latent Discovery (CDLD), a deep learning framework that explicitly models patient–provider interactions to improve prediction of blood glucose levels. By leveraging a real-world intensive care unit (ICU) dataset, the model captures latent attributes of both patients and providers, thus improving forecasting accuracy. Methods: ICU patient records were obtained from the MIMIC-IV v3.0 critical care database, including approximately 5,014 instances of patient–provider interaction. The CDLD model uses a cyclic training mechanism that alternately updates patient and provider latent representations to optimize predictive performance. During preprocessing, all numeric features were normalized, and extreme glucose values were capped at 500 mg/dL to mitigate the effect of outliers. Results: CDLD outperformed conventional models, achieving a root mean square error of 0.0852 on the validation set and 0.0899 on the test set, which indicates improved generalization. The model effectively captured latent patient–provider interaction patterns, yielding more accurate glucose variability predictions than baseline approaches. Conclusion Integrating patient–provider interaction modeling into predictive frameworks can increase blood glucose prediction accuracy. The CDLD model offers a novel approach to diabetes management, potentially paving the way for artificial intelligence-driven personalized treatment strategies.

BACKGROUND/OBJECTIVES: This study investigated the relationship between adolescent food-related lifestyles and food-related media use and eating behavior in Korea. SUBJECTS/METHODS: Participants were 392 Korean adolescents, ranging in age from 12 to 18, recruited via convenience sampling. They completed a self-report questionnaire survey consisting of questions about food-related lifestyle, food-related media use, food consumption behavior, food literacy, and nutrition quotient. Data analysis was conducted using SPSS 29.0. (IBM Co., Armonk, NY, USA). RESULTS: The factor analysis of food-related lifestyles identified four factors. Based on the cluster analysis results, participants were classified into three clusters reflecting different levels of interest: high interest in food, moderate interest in food, and low interest in food. The analysis revealed significant differences between groups in food-related liestyle factors (P < 0.05). Notably, the high-interest group demonstrated proactive engagement with food-related content, a willingness to explore diverse culinary experiences, and a conscientious consideration of nutritional labeling during food purchases. In contrast, the low-interest group reported tendencies toward overeating or succumbing to stimulating food consumption post-exposure to food-related content, coupled with a disregard for nutritional labeling when making food choices. A stronger inclination toward a food-related lifestyle was positively correlated with higher levels of food literacy and nutrition quotient. CONCLUSION This study proposes that the implementation of a nutrition education program using media could effectively promote a healthy diet among adolescents with a high level of interest in their dietary habits. For adolescents with low interest in their dietary habits, it suggests that introducing an education program with a primary focus on enhancing food literacy could be beneficial in fostering a healthy diet. Our research findings provide insight for the development of tailored nutritional education programs and establishment of effective nutrition policies.

Purpose: There are clinical unmet needs in predicting therapeutic response and precise strategy for the patient with advanced biliary tract cancer (BTC). We aimed to identify genomic alterations predicting therapeutic response and resistance to gemcitabine and cisplatin (Gem/Cis)-based chemotherapy in advanced BTC. Materials and Methods: Genomic analysis of advanced BTC multi-institutional cohorts was performed using targeted panel sequencing. Genomic alterations were analyzed integrating patients’ clinicopathologic data, including clinical outcomes of Gem/Cis-based therapy. Significance of genetic alterations was validated using clinical next-generation sequencing (NGS) cohorts from public repositories and drug sensitivity data from cancer cell lines. Results: 193 BTC patients from three cancer centers were analyzed. Most frequent genomic alterations were TP53 (55.5%), KRAS (22.8%), ARID1A (10.4%) alterations, and ERBB2 amplification (9.8%). Among 177 patients with BTC receiving Gem/Cis-based chemotherapy, ARID1A alteration was the only independent predictive molecular marker of primary resistance showing disease progression for 1st-line chemotherapy in the multivariate regression model (odds ratio, 3.12; p=0.046). In addition, ARID1A alteration was significantly correlated with inferior progression-free survival on Gem/Cis-based chemotherapy in the overall patient population (p=0.033) and in patients with extrahepatic cholangiocarcinoma (CCA) (p=0.041). External validation using public repository NGS revealed that ARID1A mutation was a significant predictor for poor survival in BTC patients. Investigation of multi-OMICs drug sensitivity data from cancer cell lines revealed that cisplatin-resistance was exclusively observed in ARID1A mutant bile duct cancer cells. Conclusion Integrative analysis with genomic alterations and clinical outcomes of the first-line Gem/Cis-based chemotherapy in advanced BTC revealed that patients with ARID1Aalterations showed a significant worse clinical outcome, especially in extrahepatic CCA. Well-designed prospective studies are mandatory to validate the predictive role of ARID1Amutation.

Purpose: This study evaluated the validity and reliability of Shively and colleagues’ self-efficacy for HIV disease management skills (HIVSE) among Korean participants. Methods: The original HIV-SE questionnaire, comprising 34 items, was translated into Korean using a translation and back-translation process. To enhance clarity and eliminate redundancy, the author and expert committee engaged in multiple discussions and integrated two items with similar meanings into a single item. Further, four HIV nurse experts tested content validity. Survey data were collected from 227 individuals diagnosed with HIV from five Korean hospitals. Construct validity was verified through confirmatory factor analysis. Criterion validity was evaluated using Pearson’s correlation coefficients with the new general self-efficacy scale. Internal consistency reliability and test-retest were examined for reliability. Results: The Korean version of HIV-SE (K-HIV-SE) comprises 33 items across six domains: “managing depression/mood,” “managing medications,” “managing symptoms,” “communicating with a healthcare provider,” “getting support/help,” and “managing fatigue.” The fitness of the modified model was acceptable (minimum value of the discrepancy function/degree of freedom = 2.49, root mean square error of approximation = .08, goodnessof-fit index = .76, adjusted goodness-of-fit index = .71, Tucker-Lewis index = .84, and comparative fit index = .86). The internal consistency reliability (Cronbach’s α = .91) and test-retest reliability (intraclass correlation coefficient = .73) were good. The criterion validity of the K-HIV-SE was .59 (p < .001). Conclusion This study suggests that the K-HIV-SE is useful for efficiently assessing self-efficacy for HIV disease management.

Cerebrospinal fluid (CSF) leakage may cause intracranial hypotension and postural headache. Secondary intracranial hypotension may result from an iatrogenic dural puncture or traumatic injury associated with pain procedures. Case: A 45-year-old male developed a headache 26 days after spinal pain procedure. Headache was characterized as postural, worsening with standing or sitting and improving while lying down. The pain did not resolve despite the administration of oral and intravenous analgesics. A spinal magnetic resonance imaging revealed epidural venous congestion and a suspicious CSF leak around the left L4/5 level. The patient received an epidural blood patch (EBP), the headache improved dramatically, and the patient was discharged. Conclusions: Delayed postural headaches may not be directly related to pain management. Nevertheless, intracranial hypotension related to pain management should be suspected even in this case. If confirmed, quickly applying an EBP is an effective treatment option.

Purpose: To investigate the clinical features of non-affected fellow eyes in patients with unilateral facial port-wine stain (PWS) and ipsilateral secondary glaucoma. Methods: We performed a retrospective analysis of the medical records of 35 patients with unilateral facial PWS glaucoma and those of controls (35 subjects without both facial PWS and glaucoma) between September 1996 and May 2020. We noted patients’ age at the glaucoma diagnosis (for unilateral facial PWS glaucoma patients) or at the initial examination (for controls), cup-to-disc ratio (CDR), and intraocular pressure (IOP). We compared the clinical features between the glaucoma-free eyes in patients with unilateral facial PWS glaucoma and the controls. Results: The mean age at the glaucoma diagnosis for unilateral facial PWS glaucoma patients was 0.56 ± 0.99 years (range, 0.08-4). The mean IOP of the glaucoma-free eyes was 16.68 ± 5.73 mmHg (range, 9-22.9), and the mean CDR was 0.37 ± 0.14 (range, 0.15-0.80) at glaucoma diagnosis. The mean IOP of the glaucoma-free eyes was 14.14 ± 6.29 mmHg (range, 8.1-26.7), and the mean CDR was 0.37 ± 0.12 (range, 0.26-0.82) at final examination. When comparing glaucoma-free eyes of the unilateral facial PWS glaucoma patients with the control group (mean age, 11.2 ± 7.4 years), the mean CDR was significantly greater (0.37 ± 0.12 vs. 0.30 ± 0.08; p = 0.014) but there was no significant difference in the mean IOP (14.14 ± 6.29 mmHg vs. 14.57 ± 2.49 mmHg; p = 0.712). Conclusions The glaucoma-free eyes of unilateral facial PWS glaucoma patients showed greater CDR compared to the non-facial PWS and non-glaucoma controls. Additional longitudinal studies are needed to investigate the clinical course of those eyes, whether the risk of developing glaucoma is increased.