Large language models (LLMs) are seen to have tremendous potential in advancing medical diagnosis recently, particularly in dermatological diagnosis, which is a very important task as skin and subcutaneous diseases rank high among the leading contributors to the global burden of nonfatal diseases. Here we present SkinGPT-4, which is an interactive dermatology diagnostic system based on multimodal large language models. We have aligned a pre-trained vision transformer with an LLM named Llama-2-13b-chat by collecting an extensive collection of skin disease images (comprising 52,929 publicly available and proprietary images) along with clinical concepts and doctors’ notes, and designing a two-step training strategy. We have quantitatively evaluated SkinGPT-4 on 150 real-life cases with board-certified dermatologists. With SkinGPT-4, users could upload their own skin photos for diagnosis, and the system could autonomously evaluate the images, identify the characteristics and categories of the skin conditions, perform in-depth analysis, and provide interactive treatment recommendations.

Cardiometabolic multimorbidity (CMM), defined as the coexistence of two or more cardiometabolic diseases, represents a growing challenge for population health and health systems worldwide. Insulin resistance and atherogenic dyslipidemia are key metabolic disturbances underlying CMM, yet their joint contribution to CMM development remains incompletely understood. A composite index integrating the triglyceride-glucose (TyG) index and the non-HDL-cholesterol-to-HDL-cholesterol ratio (NHHR), termed TyG-NHHR, was developed to examine its association with incident CMM in two population-based cohorts. Prospective data from the China Health and Retirement Longitudinal Study (2011-2018) and the English Longitudinal Study of Ageing (2008-2014) were analysed. After excluding participants with prevalent CMM or missing data, 9,297 adults aged 45 years and older were included. TyG-NHHR was calculated as TyG multiplied by NHHR. Incident CMM was defined as the new onset of at least two cardiometabolic conditions (heart disease, stroke, hypertension, or diabetes) during follow-up. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs), adjusting for sociodemographic characteristics, lifestyle factors, body mass index, and medication use. Baseline, cumulative, and trajectory patterns of TyG-NHHR were examined, with dose-response relationships assessed using restricted cubic splines. Joint effect and bidirectional mediation analyses were conducted to explore interrelationships between TyG and NHHR. During a median follow-up of 7 years, 1,851 participants (19.9%) developed CMM. Higher TyG-NHHR levels were consistently associated with increased CMM risk. Each standard deviation increase in baseline TyG-NHHR was associated with a 13% higher risk of CMM (HR 1.13, 95% CI 1.09-1.17). Participants in the highest quartile had a substantially elevated risk compared with those in the lowest quartile (HR 1.56, 95% CI 1.33-1.83), with stronger associations observed for cumulative exposure and persistently high trajectories. A linear dose-response relationship was identified. Joint-effect analysis indicated overlapping metabolic pathways between TyG and NHHR. Mediation analysis indicated that insulin resistance accounted for a substantial proportion of the association between atherogenic dyslipidemia and CMM, whereas the reverse pathway was limited. Findings were consistent across subgroups and sensitivity analyses. TyG-NHHR is a simple, routinely derived metabolic index associated with the risk of developing cardiometabolic multimorbidity in middle-aged and older adults across diverse populations. Integrated metabolic indicators such as TyG-NHHR may support early identification of individuals at elevated risk of CMM and inform population-level prevention strategies, particularly in resource-limited settings.

Cardio-Kidney-Metabolic (CKM) syndrome is a prevalent systemic disorder, but evidence linking it to air pollution is scarce. We investigated the associatio

We aimed to derive a new high-sensitivity cardiac troponin I Access (hs-cTnI-Access) 0/1h-algorithm suitable for Chinese patients. In the multi-centre study, we enrolled 369 patients with suspected non-ST-segment elevation myocardial infarction (NSTEMI). Following an adjustment, AnZhen hs-cTnI-Access 0/1h-algorithm was established with rule-in cut-off values of 0 h ≥ 50 pg/mL or Δ1h ≥ 18.0 pg/mL and rule-out cut-off values of 0 h < 3 pg/mL or 0 h < 4 pg/mL and Δ1h < 3.0 pg/mL. These revised cutoffs presented a negative predictive value (NPV) of 99.46% (95% CI,97.02–99.91%) and a sensitivity of 98.81% (95% CI,93.56–99.79%), and also resulted in a positive predictive value (PPV) of 86.42% (95% CI,77.30–92.24%) and a specificity of 96.14% (95% CI,93.22–97.83%), both of which were superior to those of the ESC 0/1 h algorithm. The new hs-cTnI-Access 0/1 h-algorithm adjusted for Chinese data is promising for improving both safety and efficacy in Chinese patients.