Purpose: Early identification of patients at risk for acute respiratory failure (ARF) could help clinicians devise preventive strategies. Analyzing biosignals with artificial intelligence (AI) can uncover hidden information and variability within time series. We aimed to develop and validate AI models to predict ARF within 72 h after emergency department admission, primarily using highresolution biosignals collected within 4 h of arrival. Materials and Methods: Our AI model, built on convolutional recurrent neural networks, combines biosignal feature extraction and sequence modeling. The model was developed and internally validated with data from 5284 admissions [1085 (20.5%) positive for ARF], and externally validated using data from 144 admissions [7 (4.9%) positive for ARF] from another institution. We defined ARF as the application of advanced respiratory support devices. Results: Our AI model performed well in predicting ARF, achieving area under the receiver operating characteristic curve (AUROC) of 0.840 and 0.743 in internal and external validations, respectively. It outperformed the Modified Early Warning Score (MEWS) and XGBoost models built only with clinical variables. High predictive ability for mortality was observed, with AUROC up to 0.809. A 10% increase in AI prediction scores was associated with 1.44-fold and 1.42-fold increases in ARF risk and mortality risk, respectively, even after adjusting for MEWS and demographic variables. Conclusion Our AI model demonstrates high predictive accuracy and significant associations with clinical outcomes. Our AI model has the potential to promptly aid in triage decisions. Our study shows that using AI to analyze biosignals advances disease detection and prediction.

Background: This study aimed to determine whether a shorter high-dose rifampicin regimen is non-inferior to the standard 6-month tuberculosis regimen. Methods: This multicenter, randomized, open-label, non-inferiority trial enrolled participants with respiratory specimen positivity by Xpert MTB/RIF assay or Mycobacterium tuberculosis culture without rifampicin-resistance. Participants were randomized at 1:1 to the investigational or control group. The investigational group received high-dose rifampicin (30 mg/kg/day), isoniazid, and pyrazinamide until culture conversion, followed by high-dose rifampicin and isoniazid for 12 weeks. The control group received the standard 6-month regimen. The primary outcome was the rate of unfavorable outcomes at 18 months post-randomization. The non-inferiority margin was set at <6% difference in unfavorable outcomes rates. The study is registered with ClinicalTrials.gov (NCT04485156) Results: Between 4 November 2020 and 3 January 2022, 76 participants were enrolled. Of these, 58 were included in the modified intention-to-treat analysis. Unfavorable outcomes occurred in 10 (31.3%) of 32 in the control group and 10 (38.5%) of 26 in the investigational group. The difference was 7.2% (95% confidence interval, ∞ to 31.9%), failing to prove non-inferiority. Serious adverse events and grade 3 or higher adverse events did not differ between the groups. Conclusion The shorter high-dose rifampicin regimen failed to demonstrate non-inferiority but had an acceptable safety profile.

Purpose: Early identification of patients at risk for acute respiratory failure (ARF) could help clinicians devise preventive strategies. Analyzing biosignals with artificial intelligence (AI) can uncover hidden information and variability within time series. We aimed to develop and validate AI models to predict ARF within 72 h after emergency department admission, primarily using highresolution biosignals collected within 4 h of arrival. Materials and Methods: Our AI model, built on convolutional recurrent neural networks, combines biosignal feature extraction and sequence modeling. The model was developed and internally validated with data from 5284 admissions [1085 (20.5%) positive for ARF], and externally validated using data from 144 admissions [7 (4.9%) positive for ARF] from another institution. We defined ARF as the application of advanced respiratory support devices. Results: Our AI model performed well in predicting ARF, achieving area under the receiver operating characteristic curve (AUROC) of 0.840 and 0.743 in internal and external validations, respectively. It outperformed the Modified Early Warning Score (MEWS) and XGBoost models built only with clinical variables. High predictive ability for mortality was observed, with AUROC up to 0.809. A 10% increase in AI prediction scores was associated with 1.44-fold and 1.42-fold increases in ARF risk and mortality risk, respectively, even after adjusting for MEWS and demographic variables. Conclusion Our AI model demonstrates high predictive accuracy and significant associations with clinical outcomes. Our AI model has the potential to promptly aid in triage decisions. Our study shows that using AI to analyze biosignals advances disease detection and prediction.

Background: We hypothesized that intranasal combination of dexmedetomidine (2 μg/kg) and ketamine (3 mg/kg) (IN DEXKET) improves the success rate of sedation in pediatric patients compared with chloral hydrate (CH; 50 mg/kg). Methods: This prospective, two-center, single-blinded, randomized controlled trial involved 136 pediatric patients (aged < 7 years) requiring procedural sedation. The participants were randomized to receive CH or IN DEXKET via a mucosal atomizer device. The primary outcome was the success rate of sedation (Pediatric Sedation State Scale, scores 1–3) within 15 min. The secondary outcomes included sedation failure at 30 min and overall complications of first-attempt sedation. Results: After excluding eight patients, 128 were included (CH = 66, IN DEXKET = 62). IN DEXKET showed a similar sedation success rate (75.8% [47/62] vs. 66.7% [44/66]; P = 0.330) but a lower complication rate (3.2% [2/62] vs. 16.7% [11/66]; P = 0.017) than CH. In the subgroup analysis for patients aged < 1 year, IN DEXKET showed a reduced complication rate than CH (2.6% [1/38] vs. 22.9% [8/35]; P = 0.012). In the subgroup analysis of children aged 1–7 years, IN DEXKET showed a higher sedation success rate within 15 min (79.2% [19/24] vs. 51.6% [16/31]; P = 0.049) and a lower sedation failure after 30 min (0% vs. 29.0% [9/31]; P = 0.003) than CH. Conclusions The intranasal combination of dexmedetomidine (2 μg/kg) and ketamine (3 mg/kg) is a safe and effective alternative to CH (50 mg/kg) for sedation in pediatric patients aged < 7 years.

Purpose: This study aimed to evaluate the impact of declining surgical residency program enrollment on patient care and outcomes in colorectal cancer surgeries. Methods: This retrospective observational study included 676 patients (410 males; median age: 69 years) who underwent colorectal cancer surgery at Pusan National University Hospital between January 2018 and June 2024. Patients were divided into Group A (before December 31, 2023; with residents) and Group B (after January 1, 2024; without residents). All surgeries were performed by a single attending surgeon. Results: Preoperative variables were comparable between groups. Group A had more emergency and open surgeries, and a higher proportion of advanced-stage cancers. Overall complication rates were similar, but Group B had a longer hospital stay (9.72 days vs. 11.95 days). Specific complications such as anastomotic leakage and surgical site infections differed significantly. The overall number of surgical procedures declined markedly in 2024 compared to 2018 (77.1% vs. 49.9%). Conclusion The absence of residents did not increase overall complication rates but was associated with longer hospital stays and shifts in clinical practice. Greater reliance on attending surgeons contributed to more defensive decision-making and conservative patient management. Addressing these issues requires systemic reforms, including multidisciplinary collaboration and legal protections to improve surgical care.

Background: Skipping breakfast is associated with an increased risk of chronic inflammatory diseases. This study aimed to examine the association between breakfast-eating habits and inflammation, using high-sensitivity C-reactive protein (hs-CRP) as a marker. Methods: A total of 4,000 Korean adult males with no history of myocardial infarction, angina, stroke, diabetes, rheumatoid arthritis, cancer, or current smoking were included. Data from the 2016–2018 Korea National Health and Nutrition Examination Survey were used for analysis. The frequency of breakfast consumption was assessed through a questionnaire item in the dietary survey section asking participants about their weekly breakfast consumption routines over the past year. Participants were categorized into two groups, namely “0–2 breakfasts per week” and “3–7 breakfasts per week”; hs-CRP concentrations were measured through blood tests. Results: Comparing between the “infrequent breakfast consumption (0–2 breakfasts per week)” and “frequent breakfast consumption (3–7 breakfasts per week)” groups, the mean hs-CRP was found to be significantly higher in the “infrequent breakfast consumption” group, even after adjusting for age, body mass index, physical activity, alcohol consumption, systolic blood pressure, blood pressure medication, fasting blood glucose, and triglycerides (mean hs-CRP: frequent breakfast consumption, 1.36±0.09 mg/L; infrequent breakfast consumption, 1.17±0.05 mg/L; P-value=0.036). Conclusion Less frequent breakfast consumption was associated with elevated hs-CRP levels. Further large-scale studies incorporating adjusted measures of daily eating patterns as well as food quality and quantity are required for a deeper understanding of the role of breakfast in the primary prevention of chronic inflammatory diseases.

Purpose: This study aimed to analyze whether the occurrence of complications increases if radiotherapy (RT) is administered after breast reconstructive surgery using implants. Methods: This retrospective study included 80 patients who underwent breast reconstruction using implants, of which 16 (20.0%) underwent RT. Most patients underwent conventional fractionated RT (n = 13), and hypofractionated RT was performed in 3 patients. Most patients (n = 51, 63.8%) underwent delayed reconstruction, which involved implant replacement after tissue expander insertion. Only 29 patients (36.3%) underwent immediate reconstruction simultaneously with breast cancer surgery. Results: The median postoperative follow-up was 39.9 months (range, 8.7–120.3 months). Complications occurred in 18 (22.5%); infectionecrosis (n = 8), leakage/rupture (n = 8), and capsular contracture (n = 2). Infectionecrosis is common in patients undergoing RT. Complications occurred in 4 patients (25.0%) who received RT and 14 (21.9%) who did not receive RT, and complications did not significantly increase with RT (P = 0.511). There was no overall difference in complications between the immediate (4 of 29) and delayed (14 of 51) reconstruction groups (P = 0.129). Nine patients underwent reoperation because of complications; 3 (18.8%) received RT and 6 (9.4%) did not receive RT. The reoperation rate did not increase significantly with RT (P = 0.254). There were 3 cases of recurrence, and patients who received RT had no recurrence. Conclusion RT did not significantly increase the complication or reoperation rates if reconstructive surgery was performed using implants. Therefore, RT should be performed in patients at a high risk of recurrence.

Purpose: The purpose of this study was to develop an approach to alleviate the discomfort caused by sandbag compression after a bone marrow examination. This research examined the effects of applying a pressure hemostasis band on bleeding, pain, and discomfort at the bone marrow examination site. Methods: This study was conducted with a nonequivalent control group non-synchronized design. For 74 patients under evaluation who underwent bone marrow examination, sandbag compression was applied to the examination site in the control group (n=37), and a pressure hemostasis band was applied to the intervention group (n=37). In both groups, absolute bed rest was performed for two hours, and bleeding, pain, and discomfort at the examination site were measured. Results: After two hours of the bone marrow examination, there was no difference in bleeding on the gauze between the two groups (F=0.59, p=.444). Bleeding occurred in three patients in the intervention group and six in the control group (χ 2 =1.14, p=.479), with no cases of hematoma detected in either group. One hour post-examination, the control group experienced significantly higher pain (F=5.45, p=.022) and discomfort (F=5.68, p=.020) than the intervention group. However, pain and discomfort levels were similar between groups after two hours. Conclusion Compared to the sandbag compression group, the band application group showed no difference in bleeding and experienced less pain and discomfort at the examination site. This confirms that the pressure hemostasis band is a suitable alternative to sandbag compression in post-examination care.

The common data model (CDM) has found widespread application in healthcare studies, but its utilization in cancer research has been limited. This article describes the development and implementation strategy for Cancer Clinical Library Databases (CCLDs), which are standardized cancer-specific databases established under the Korea-Clinical Data Utilization Network for Research Excellence (K-CURE) project by the Korean Ministry of Health and Welfare. Fifteen leading hospitals and fourteen academic associations in Korea are engaged in constructing CCLDs for 10 primary cancer types. For each cancer type-specific CCLD, cancer data experts determine key clinical data items essential for cancer research, standardize these items across cancer types, and create a standardized schema. Comprehensive clinical records covering diagnosis, treatment, and outcomes, with annual updates, are collected for each cancer patient in the target population, and quality control is based on six-sigma standards. To protect patient privacy, CCLDs follow stringent data security guidelines by pseudonymizing personal identification information and operating within a closed analysis environment. Researchers can apply for access to CCLD data through the K-CURE portal, which is subject to Institutional Review Board and Data Review Board approval. The CCLD is considered a pioneering standardized cancer-specific database, significantly representing Korea’s cancer data. It is expected to overcome limitations of previous CDMs and provide a valuable resource for multicenter cancer research in Korea.

Objectives: . Salivary gland tumors present a diagnostic challenge, with preoperative false-negative results frequently leading to an unexpected diagnosis of malignancy after parotidectomy. This study was conducted to explore the clinical utility of preoperative core needle biopsy (CNB) in diagnosing malignancies before primary parotidectomy and to assess the prognostic implications of CNB for parotid gland cancers. Methods: . This retrospective cohort study included 615 patients who underwent preoperative CNB and parotidectomy for primary parotid tumors from 2003 to 2023 at a tertiary referral hospital. Among these patients, 102 who were diagnosed with primary parotid malignancy following parotidectomy were examined regarding survival outcomes. Disease-free survival (DFS) and predictive factors were assessed through univariable and multivariable analyses. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy of CNB were determined. These metrics were then compared to those of a separate cohort of 547 patients who underwent ultrasound-guided fine needle aspiration (FNA) and parotidectomy within the same timeframe. Results: . In the CNB group, the 5-year predicted DFS was 86.9% (95% CI, 79.2%–95.3%). Multivariable analysis identified male sex (hazard ratio [HR], 8.48; 95% CI, 1.05–68.76) and a CNB finding of malignancy (HR, 8.20; 95% CI, 1.01–66.15) as factors significantly associated with decreased DFS. CNB demonstrated significantly higher sensitivity (89.0%; 95% CI, 81.2%–94.4%) and NPV (97.6%; 95% CI, 95.7%–98.8%) compared to FNA, which had a sensitivity of 45.2% (95% CI, 33.5%–57.3%) and an NPV of 90.8% (95% CI, 87.7%–93.3%). Conclusion . Preoperative CNB may be predictive of parotid cancer prognosis. Patients receiving a diagnosis of malignancy on preoperative CNB demonstrated a worse prognosis compared to those with a “less-than-malignant” diagnosis. Additionally, CNB exhibited higher sensitivity than FNA in identifying malignancies of the parotid gland.