Objectives: This research aimed to develop a retrieval-augmented generation (RAG) based large language model (LLM) system that offers personalized and reliable responses to a wide range of concerns raised by Korean adolescents. Our work focuses on building a culturally reflective dataset and on designing and validating the system’s effectiveness by comparing the answer quality of RAG-based models with non-RAG models. Methods: Data were collected from the NAVER Knowledge iN platform, concentrating on posts that featured adolescents’ questions and corresponding expert responses during the period 2014–2024. The dataset comprises 3,874 cases, categorized by key negative emotions and the primary sources of worry. The data were processed to remove irrelevant or redundant content and then classified into general and detailed causes. The RAG-based model employed FAISS for similarity-based retrieval of the top three reference cases and used GPT-4o mini for response generation. The responses generated with and without RAG were evaluated using several metrics. Results: RAG-based responses outperformed non-RAG responses across all evaluation metrics. Key findings indicate that RAG-based responses delivered more specific, empathetic, and actionable guidance, particularly when addressing complex emotional and situational concerns. The analysis revealed that family relationships, peer interactions, and academic stress are significant factors affecting adolescents’ worries, with depression and stress frequently co-occurring. Conclusions This study demonstrates the potential of RAG-based LLMs to address the diverse and culture-specific worries of Korean adolescents. By integrating external knowledge and offering personalized support, the proposed system provides a scalable approach to enhancing mental health interventions for adolescents. Future research should concentrate on expanding the dataset and improving multiturn conversational capabilities to deliver even more comprehensive support.

Objectives: This research aimed to develop a retrieval-augmented generation (RAG) based large language model (LLM) system that offers personalized and reliable responses to a wide range of concerns raised by Korean adolescents. Our work focuses on building a culturally reflective dataset and on designing and validating the system’s effectiveness by comparing the answer quality of RAG-based models with non-RAG models. Methods: Data were collected from the NAVER Knowledge iN platform, concentrating on posts that featured adolescents’ questions and corresponding expert responses during the period 2014–2024. The dataset comprises 3,874 cases, categorized by key negative emotions and the primary sources of worry. The data were processed to remove irrelevant or redundant content and then classified into general and detailed causes. The RAG-based model employed FAISS for similarity-based retrieval of the top three reference cases and used GPT-4o mini for response generation. The responses generated with and without RAG were evaluated using several metrics. Results: RAG-based responses outperformed non-RAG responses across all evaluation metrics. Key findings indicate that RAG-based responses delivered more specific, empathetic, and actionable guidance, particularly when addressing complex emotional and situational concerns. The analysis revealed that family relationships, peer interactions, and academic stress are significant factors affecting adolescents’ worries, with depression and stress frequently co-occurring. Conclusions This study demonstrates the potential of RAG-based LLMs to address the diverse and culture-specific worries of Korean adolescents. By integrating external knowledge and offering personalized support, the proposed system provides a scalable approach to enhancing mental health interventions for adolescents. Future research should concentrate on expanding the dataset and improving multiturn conversational capabilities to deliver even more comprehensive support.

Objectives: This research aimed to develop a retrieval-augmented generation (RAG) based large language model (LLM) system that offers personalized and reliable responses to a wide range of concerns raised by Korean adolescents. Our work focuses on building a culturally reflective dataset and on designing and validating the system’s effectiveness by comparing the answer quality of RAG-based models with non-RAG models. Methods: Data were collected from the NAVER Knowledge iN platform, concentrating on posts that featured adolescents’ questions and corresponding expert responses during the period 2014–2024. The dataset comprises 3,874 cases, categorized by key negative emotions and the primary sources of worry. The data were processed to remove irrelevant or redundant content and then classified into general and detailed causes. The RAG-based model employed FAISS for similarity-based retrieval of the top three reference cases and used GPT-4o mini for response generation. The responses generated with and without RAG were evaluated using several metrics. Results: RAG-based responses outperformed non-RAG responses across all evaluation metrics. Key findings indicate that RAG-based responses delivered more specific, empathetic, and actionable guidance, particularly when addressing complex emotional and situational concerns. The analysis revealed that family relationships, peer interactions, and academic stress are significant factors affecting adolescents’ worries, with depression and stress frequently co-occurring. Conclusions This study demonstrates the potential of RAG-based LLMs to address the diverse and culture-specific worries of Korean adolescents. By integrating external knowledge and offering personalized support, the proposed system provides a scalable approach to enhancing mental health interventions for adolescents. Future research should concentrate on expanding the dataset and improving multiturn conversational capabilities to deliver even more comprehensive support.

PURPOSE: Emergency physicians usually perform ultrasound exams for patients with multiple injuries. Extended focused assessment with sonography for trauma (eFAST) can reveal injuries from chest to the abdomen immediately. However, in Korea, the curriculums of medical schools do not currently include eFAST. We have devised a study to assess the feasibility of the eFAST exam in medical school students. METHODS: This study was conducted in students in their fifth year out of six years of medical school, over 11 weeks. Four Emergency Medicine specialists trained the students over 4 hours, tested the students, and conducted a questionnaire. RESULTS: Average age of students was 25.9+/-2.6 years, and 24 were male and 20 were female. Mean success rate of 17 components on the eFAST exam was 95.9% (94.6%-97.4%). The success rate of transverse view of aorta, transverse view of bladder, lung sliding sign, and sea-shore sign of both anterior chest walls was 100%, scanning the spleen and attaching the probe to a body surface were 75%, 86.3%. Total time consumption was 449.0+/-22.2 seconds. The questionnaires showed that the eFAST exam for the splenorenal recess, spleen, and left upper quadrant lung was difficult. CONCLUSION: We found that medical students in Korea could perform the eFAST exam by themselves after 4 hours education. In the future, these findings can be helpful in development of an eFAST education program for medical students.
Abdomen ; Aorta ; Curriculum ; Education ; Emergencies ; Emergency Medicine ; Female ; Humans ; Korea ; Lung ; Male ; Multiple Trauma ; Surveys and Questionnaires ; Schools, Medical* ; Specialization ; Spleen ; Students, Medical ; Thorax ; Ultrasonography ; Urinary Bladder

PURPOSE: The purpose of this study was to report medical care activities of Disaster Medical Assistance Team (DMAT) and medical facilities that responded to the Goyang Bus Terminal fire on May 26, 2014, and to draw improvement of the current disaster medical response system. METHODS: We retrospectively reviewed emergency medical service (EMS) run sheet and medical records of patients who visited the emergency department the day of the fire. We also interviewed the officials involved in disaster response. RESULTS: A total of 73 patients participated in this study. Among them, 9 were classified as Emergency, 60 as Non-Emergency, and 4 as Death on arrival (DOA). Fifty one patients visited the nearest hospital, and 17 patients were transported by EMS. DMAT arrived at the scene in 58 minutes, however there was little medical activity. CONCLUSION: Initial Triage and distribution of patients was rather inadequate and DMAT arrived late. For the future, we recommend constant training of the paramedics and leaders of 119, and to mend DMAT requesting and response system.
Allied Health Personnel ; Disasters* ; Emergencies ; Emergency Medical Services ; Emergency Service, Hospital ; Fires* ; Humans ; Mass Casualty Incidents ; Medical Assistance ; Medical Records ; Retrospective Studies ; Triage

Ataxia is presented by various etiologies, including acquired, genetic and degenerative disorders. Although hereditary ataxia is suspected when typical symptom of ataxia with concurrent is identified, it is sometimes difficult to diagnose hereditary ataxia without genetic test. Clinically, next generation sequencing technology has been developed and widely used for diagnosis of hereditary disease. Hereby, we experienced cases of genetically confirmed OPA1 mutation, which are presented with various clinical manifestations including ataxic gait and decreased visual acuity.

The rapid aging trend in South Korea will cause a growing shortage of labor and decreasing quality of the labor force. The purpose of this commentary is to recommend a health care strategy to maintain and promote the work ability of employees in an aging Korea. Strategies to promote the work ability require the collaboration of governmental agencies at the central and local levels. First, the common goal should be the reinforcement of follow-up measure in general medical examinations and the promotion of healthy lifestyles for workers. Second, collaborating activities should be performed among the Worker’s Health Center, the Health Promotion Center, and community health centers. In conclusion, health care strategies for ensuring the work ability in an aging Korea require the collaboration of governmental agencies at the central and local levels.
Aging* ; Community Health Centers ; Cooperative Behavior ; Delivery of Health Care* ; Employment ; Follow-Up Studies ; Health Promotion ; Korea* ; Life Style

OBJECTIVE: We compared the clinical and radiographic outcomes of stand-alone polyetheretherketone (PEEK) cage and Zero-Profile anchored spacer (Zero-P) for single level anterior cervical discectomy and fusion (ACDF). METHODS: We retrospectively reviewed 121 patients who underwent single level ACDF within 2 years (Jan 2011-Jan 2013) in a single institute. Total 50 patients were included for the analysis who were evaluated more than 2-year follow-up. Twenty-nine patients were allocated to the cage group (m : f=19 : 10) and 21 for Zero-P group (m : f=12 : 9). Clinical (neck disability index, visual analogue scale arm and neck) and radiographic (Cobb angle-segmental and global cervical, disc height, vertebral height) assessments were followed at pre-operative, immediate post-operative, post-3, 6, 12, and 24 month periods. RESULTS: Demographic features and the clinical outcome showed no difference between two groups. The change between final follow-up (24 months) and immediate post-op of Cobb-segmental angle (p=0.027), disc height (p=0.002), vertebral body height (p=0.033) showed statistically better outcome for the Zero-P group than the cage group, respectively. CONCLUSION: The Zero-Profile anchored spacer has some advantage after cage for maintaining segmental lordosis and lowering subsidence rate after single level anterior cervical discectomy and fusion.
Animals ; Arm ; Body Height ; Diskectomy* ; Follow-Up Studies ; Humans ; Kyphosis ; Lordosis ; Retrospective Studies

The sacral spinal epidural space is an uncommon site for primary malignant lymphomas, presenting with symptoms associated with cauda equina compression. Especially, lumbo-sacral epidural lymphoma has been reported to be very rare. We present a rare case of 29-year-old male with sacral spinal epidural malignant lymphoma. The patient complained of tingling sensation in his buttocks that was radiating to his calf. The neurological examination was normal. Magnetic resonance imaging (MRI) with contrast showed a well-defined extradural mass lesion at the mid L5 to mid S2 level. The lesion was iso- to hypointense on T1 and T2 weighted images and showed homogenous enhancement and a focal enhancement in the L5 vertebral body on post-contrast images. The patient underwent a L5-S2 laminectomy and subtotal excision of the lesion. Intra-operatively, the lesion was extradural and not densely adherent to the dura; the lesion was friable, not firm, fleshy, brownish and hypervascular. The histologic diagnosis was grade 2 non-Hodgkin's follicular lymphoma. Even though the primary spinal epidural non-Hodgkin's lymphoma is a very rare disease, clinicians should take it into consideration in the differential diagnosis of patients with spinal epidural tumor.
Adult ; Buttocks ; Cauda Equina ; Diagnosis ; Diagnosis, Differential ; Epidural Neoplasms ; Epidural Space* ; Humans ; Laminectomy ; Lymphoma* ; Lymphoma, Follicular ; Lymphoma, Non-Hodgkin ; Magnetic Resonance Imaging ; Male ; Neurologic Examination ; Rare Diseases ; Sensation

PURPOSE: Falling is one of the most common causes of injury for preschool children. Here, we aim to identify the characteristics and risk factors of injuries by falling-down in preschool children. METHODS: Between January 2010 and December 2011, we enrolled patients under the age of 7 years, who were injured by falling down and visited an urban regional emergency center. We retrospectively surveyed the medical record of these patients, including age, sex, place and height of fall, type of floor, guardian's witness, traumatic brain injury (TBI), and fracture of extremities. RESULTS: The odds ratios sex (male), age (under 2 years old), height of fall, type of floor (hard), and guardian's witness (presence) that resulted in TBI were 1.35 (95% confidence interval [Cl], 0.72?2.55; p=0.352), 3.83 (95% Cl, 1.78?8.65; p<0.05), 6.38 (95% Cl, 3.27–12.44;p<0.05), 3.58 (95% Cl, 0.47–27.30; p=0.218), and 1.47 (95% Cl, 0.63–3.43; p=0.377), respectively. The odds ratios sex (male), age (over 2 years old), height of fall, type of floor (soft), and guardian's witness (absence) that resulted in fractures were 1.19 (95% Cl, 0.78–1.81; p=0.433), 3.10 (95%Cl, 1.99–4.84; p<0.05), 1.98 (95%Cl, 1.19–3.29; p<0.05), 2.41 (95% Cl, 1.29–4.54; p<0.05), and 1.15 (95%Cl, 0.72–1.85, p=0.554), respectively. CONCLUSION: In preschool children who experienced an injury from falling down, TBI was increased with younger patients and higher height of fall, but it was not related with patient's sex, type of floor, and guardian's witness; conversely, fractures were increased with older patients, higher height of fall, and soft floor, but not related with patient's sex and guardian's witness.
Accidental Falls ; Brain Injuries ; Child* ; Child, Preschool ; Emergencies ; Extremities ; Humans ; Medical Records ; Odds Ratio ; Retrospective Studies ; Risk Factors*