Mennonite Christian Hospital

Background: Extranodal extension (ENE) is a critical prognostic factor in head and neck squamous cell carcinoma (HNSCC) and was incorporated into the AJCC eighth-edition staging system. However, the concordance between clinical (cENE) and pathological (pENE) ENE remains poorly understood in real-world practice. Methods: We conducted a retrospective analysis using Taiwan Cancer Registry (TCR) long-form data from 2018 to 2022, focusing on four major HNSCC sites (oral cavity, oropharynx, hypopharynx, and larynx). The diagnostic gap was defined as the difference between pENE and cENE positivity rates. Results: Among 29,830 patients, a persistent diagnostic gap was observed across all sites: laryngeal (20.8%), hypopharyngeal (20.4%), oropharyngeal (11.5%), and oral cavity (9.9%). For oral cavity cancer, the gap did not narrow over the 5-year period (p = 0.9788). Furthermore, in oral cavity cancer, medical centers demonstrated a larger gap than non-medical centers (10.5% vs. 8.4%), a phenomenon we term the "Quality-Gap Paradox". Conclusions: A significant diagnostic gap persists in HNSCC, highlighting the limitations of current imaging. The Quality-Gap Paradox, observed in oral cavity cancer, suggests this is driven by a complex interplay of factors including superior pathological detection in high-volume centers. Our findings underscore the need for advanced, personalized risk-stratification tools to bridge this gap and improve patient management.

To examine global publication trends, thematic developments, and emerging priorities in research on physical restraint from 2005 to 2024. The current bibliometric analysis was conducted using original research published in English between 2005 and 2024, retrieved from the Science Citation Index Expanded database. Analyses of publication trends, institutional and geographic productivity, and author keyword networks were performed using Bibliometrix. A total of 1,154 articles were identified, with a 5.5% annual growth rate. Long-term care and dementia, intensive care sedation and delirium, psychiatric coercion, and health care professionals' knowledge and attitudes emerged as dominant themes. Thematic evolution revealed a shift from sedation protocols and agitation management to ethical issues, and a recent emphasis on dementia care, behavioral assessments, and interventions. The field has transitioned toward ethical reflection and interdisciplinary restraint reduction. Future priorities include exploring non-institutional contexts and integrating technology for restraint-minimizing care models.

Background/Objectives: Interpretation of chest X-rays (CXRs) requires accurate identification of lesion presence, diagnosis, location, size, and number to be considered complete. However, the effectiveness of large language models with vision capabilities (LLMs) in performing these tasks remains uncertain. This study aimed to evaluate the image-only interpretation performance of LLMs in the absence of clinical information. Methods: A total of 247 CXRs covering 13 diagnostic categories, including pulmonary edema, cardiomegaly, lobar pneumonia, and other conditions, were evaluated using Gemini 1.0, Gemini 1.5 Pro, GPT-4 Turbo, and GPT-4o. The text outputs generated by the LLMs were evaluated at two levels: (1) primary diagnosis accuracy across the 13 predefined diagnostic categories, and (2) identification of key imaging features described in the generated text. Primary diagnosis accuracy was assessed based on whether the model correctly identified the target diagnostic category and was classified as fully correct, partially correct, or incorrect according to predefined clinical criteria. Non-diagnostic imaging features, such as posteroanterior and anteroposterior (PA/AP) views, side markers, foreign bodies, and devices, were recorded and analyzed separately rather than being incorporated into the primary diagnostic scoring. Results: When fully and partially correct responses were treated as successful detections, vLLMs showed higher sensitivity for large, bilateral, multiple lesions and prominent devices, including acute pulmonary edema, lobar pneumonia, multiple malignancies, massive pleural effusions, and pacemakers, all of which demonstrated statistically significant differences across categories in chi-square analyses. Feature descriptions varied among models, especially in PA/AP views and side markers, though central lines were partially recognized. Across the entire dataset, Gemini 1.5 Pro achieved the highest overall detection rate, followed by Gemini 1.0, GPT-4o, and GPT-4 Turbo. Conclusions: Although LLMs were able to identify certain diagnoses and key imaging features, their limitations in detecting small lesions, recognizing laterality, reasoning through differential diagnoses, and using domain-specific expressions indicate that CXR interpretation without textual cues still requires further improvement.

Journal of Minimally Invasive Spine Surgery & Technique (JMISST) is the official journal of the Korean Minimally Invasive Spine Surgery Society (KOMISS) for the publication of research results about minimally invasive spinal surgery (MISS), including studies in spinal endoscopic surgery, minimally invasive procedure for degenerative spine disease, pain intervention, and minimally invasive surgery for spinal fusion or spine trauma, neuroscience, neurology, molecular biology and biomechanics etc. Articles with helpful information for MISS such as reports of rare cases or technical review of special devices or equipment are also available

Diffusion models, variational autoencoders, and generative adversarial networks (GANs) are three common types of generative artificial intelligence models for image generation. Among these, GANs are the most frequently used for medical image generation and are often employed for data augmentation in various studies. However, due to the adversarial nature of GANs, where the generator and discriminator compete against each other, the training process can sometimes end with the model unable to generate meaningful images or even producing noise. This phenomenon is rarely discussed in the literature, and no studies have proposed solutions to address this issue. Such outcomes can introduce significant bias when GANs are used for data augmentation in medical image training. Moreover, GANs often require substantial computational power and storage, adding to the challenges. In this study, we used deep convolutional GANs for chest X-raygeneration, and three typical training outcomes were found. Two scenarios generated meaningful medical images and one failed to produce usable images. By analyzing the loss history during training, we observed that the regression line of the overall losses tends to diverge slowly. After excluding outlier losses, we found that the slope of the regression line within the stable loss segment indicates the optimal point to terminate training, ensuring the generation of meaningful medical images.

BackgroundHepatic inflammation, e.g., Nonalcoholic Fatty Liver Diseases (NAFLD) and the severe form of steatohepatitis (NASH), has been associated with a higher risk of MetS in the general population.AimsThis study aimed to investigate the associations of chronic hepatitis B (CHB) and fatty liver diseases with the incidence of metabolic syndrome (MetS) in young adults, which have not been verified before.ObjectiveThe associations between NAFLD, NASH, and CHB and the incidence of new-onset MetS remain inconclusive in young adults.MethodsThis cohort study included 2,614 military personnel aged 18-39 years who were free of baseline MetS in 2014 and were followed for the incidence of MetS in each annual health examination until the end of 2020. CHB was defined by the presence of the hepatitis B surface antigen with an established diagnosis history. NAFLD and NASH were defined by the ultrasound finding with an elevated alanine transaminase (27-53.9 and ≥54 U/L in men and 15-29.9 and ≥30 U/L in women, respectively). MetS was defined based on the International Diabetes Federation criteria. Multivariable Cox regression analysis was used to determine the associations between hepatitis and incident MetS.ResultsDuring a mean follow-up of 6.0 years, 582 new-onset MetS cases occurred (22.3%). NAFLD and NASH were associated with a greater risk of new-onset MetS (hazard ratios (HRs) and 95% confidence intervals: 1.47 (1.21-1.79) and 1.66 (1.16-2.39), respectively), while no association for CHB was found (HR: 1.31 (0.88-1.96)).ConclusionThis study found that NAFLD and NASH, while not CHB, were independent risk factors of new-onset MetS with adjustments for potential covariates, e.g., physical activity and fitness in young adults.

The study aimed to compare the predictive capabilities of the traditional anthropometric indices with the novel anthropometric indices for incident hypertension. Some novel anthropometric indices, e.g., the Body Roundness Index (BRI) and A Body Shape Index (ABSI) have been associated with prevalent hypertension. There are a few cohort studies that have examined the association of the novel anthropometric indices with newonset hypertension in young adults. This study included 2,448 military male and female young adults, aged 18-39 years, free of hypertension at baseline in Taiwan; they were followed for incidence of hypertension from 2014 till the end of 2020. Blood Pressure (BP) in mmHg was measured twice and averaged to verify hypertension, which was defined as systolic BP ≥130 and/or diastolic BP ≥80 or on antihypertensive medication therapy in each annual health examination. Anthropometric indices included the Body Mass Index (BMI) defined as the weight (kg)/height squared (m2), Waist Girth (WC) in cm, the Waist-to-height Ratio (WHtR), the BRI defined as 364.2 - 365.5 × {1 - [(WC/2π)/(0.5 × height)]2}0.5, as well as ABSI defined as WC/(BMI2/3 × height1/2). Multiple Cox regression analysis and Area Under the Curve (AUC) of the Receiver of Operating Characteristics (ROC) were utilized with adjustments for the baseline potential covariates to determine the association and compare the performance of various indices for incident hypertension. During a median follow-up period of 6.0 years, 920 new-onset hypertension cases (37.6%) developed. Higher BMI, WC, BRI (per each 1-unit increase) and WHtR (per each 0.1- unit increase) were associated with a greater risk of new-onset hypertension (Hazard Ratios (HRs) and 95% confidence intervals: 1.060 (1.035-1.085), 1.021 (1.011-1.030), and 1.178 (1.077-1.288), respectively), whereas there was no association between ABSI and new-onset hypertension. For the ROC, WC was observed with the greatest AUC for incident new-onset hypertension (0.661 (0.638-0.683)), followed by BMI (0.650 (0.628-0.673)), while the ABSI was found with the lowest AUC (0.544 (0.521-0.568)). Most of the anthropometric indices were associated with a higher risk of new-onset hypertension among young adults, except for ABSI. In addition, this study has suggested the traditional indices, such as WC and BMI, to be superior to the latest ones, e.g., BRI and ABSI, for the prediction of new-onset hypertension.

Difficult airway management (DAM) presents a complex array of challenges inherent in establishing and maintaining a patient's airway during medical procedures or emergencies. Consequences of compromised DAM are profound, ranging from hypoxia, and aspiration, to cardiac arrest. Despite the ongoing progress and innovation in airway management, DAM remains a significant clinical challenge. Performing a bibliometric analysis of the most cited articles in this domain offers quantitative and qualitative insights into the landscape and trend of key research; our analysis revealed an increasing focus on video laryngoscopy research over the past decade, with a notable surge in studies related to airway management during infectious disease outbreaks in recent years. We performed a bibliometric analysis of the 100 most-cited articles on DAM (Top100DAM) published between 2011 and 2022, sourced from both PubMed and Web of Science databases. Employing social network analysis, we identified key associated topics and cluster terms. Cluster terms are broad categories encompassing related topic terms. Topic terms are specific keywords or phrases found in the articles. Citation counts were estimated for each article, and cluster-specific reviews were generated using the Medical Query Expert software. The Top100DAM comprised 53 topic terms spanning 11 cluster terms. "Anesthesiology" stood out as the most frequently featured cluster, and "betacoronavirus" had the highest Impact Factor. There was a significant correlation between the counts between the inferred citation based on cluster terms and the observed citation (Pearson's R = 0.06, P = 0.02). To visualize, the network visualization and Sankey diagrams were used to illustrate the characteristics of the articles. The generated summary, utilizing natural language processing techniques, underscored the role of fiber-optic technology and video laryngoscopy in addressing DAM. During the COVID-19 pandemic, experts recommended the provision of supplemental oxygen and the prioritization of video laryngoscopy for anticipated DAM scenarios. Notably, the most cited articles predominantly concentrated on the importance of practice driven by guidelines and protocols for clinicians and anesthesiologists to manage difficult airways effectively. DAM plays a central role in anesthesiology research, with a significant emphasis on guideline development, technological advancements like video laryngoscopy, and adapting practices for infectious disease scenarios. The increased publication volume and high citation rates during the COVID-19 pandemic underscore how ongoing research and innovation have accelerated, refining DAM practices. Articles with high citation rates mostly pertain to guidelines, illustrating their substantial impact on clinical practice and the necessity for studies to base their discussions on these guidelines. Future research should focus on addressing the implementation challenges of existing guidelines and developing context-specific protocols for diverse clinical settings.

BackgroundInsulin resistance is associated with the development of hypertension, whereas there were rare studies comparing various non-insulin based insulin resistance (NI-IR) indices for the possibility of hypertension among young and middle-aged adults.MethodsThis cross-sectional study included a total of 4,080 military personnel, aged 18–50 years, without antihypertensive medications therapy in 2014. All subjects received annual health examinations for blood pressure (BP) measurements. Stage I isolated diastolic hypertension (IDH) and isolated systolic hypertension (ISH) and combined hypertension were respectively defined as systolic BP (SBP) < 130 mmHg/diastolic BP (DBP) 80–89 mmHg, SBP 130–139 mmHg/DBP < 80 mmHg, and SBP 130–139 mmHg/DBP 80–89 mmHg. The cut-off values of stage II hypertension for SBP and DBP were 140–159 mmHg and 90–99 mmHg, respectively. Four NI-IR indices included the serum triglycerides (TG) to high-density lipoprotein cholesterol (HDL-C) ratio, TyG index, Metabolic Score for IR (METS-IR) and ZJU index which were defined according to their specific formula. Multiple logistic regression analysis with adjustments for age, sex, anthropometrics, substance use, kidney function, serum uric acid, atherogenic cholesterols and physical activity was performed to determine the associations.ResultsThere were 1,024 subjects with hypertension (25.1%) in which 739 were stage I hypertension, and 285 were stage II hypertension. For total hypertension, there were an association with TyG and METS-IR indices [odds ratios (ORs) and 95% confidence intervals: 1.432 (1.215–1.688) and 1.553 (1.040–2.321), respectively]. For hypertension subtypes, TyG index was positively associated with overall, stage I, and stage II ISH [ORs: 1.447 (1.149–1.823), 1.317 (1.029–1.687), and 2.011 (1.351–2.994), respectively], while TG/HDL-C, METS-IR and ZJU indices were merely associated with stage II ISH [ORs: 1.053 (1.006–1.103), 3.001 (1.171–7.696) and 1.009 (1.000-1.017), respectively]. In addition, TyG and METS-IR indices were positively associated with stage II IDH [ORs: 1.813 (1.207–2.721) and 2.85 (1.080–7.520), respectively], and TyG index was also associated with combined hypertension [OR: 1.425 (1.007–1.833)].ConclusionAmong young and middle-aged adults, insulin resistance assessed by the four NI-IR indices was positively associated with stage II ISH, while only TyG index had a significant association for both stage II IDH and combined hypertension.