CARDIOLOGY

Researchers have developed a machine learning model, InterTAK-ML, to predict in-hospital death risks in Takotsubo syndrome (TTS) patients. The study analyzed data from 4,519 TTS cases, identifying 31 key variables. In internal validation, InterTAK-ML demonstrated high accuracy with an AUC of 0.89, sensitivity of 0.85, and specificity of 0.76. External validation yielded similar results with an AUC of 0.82, sensitivity of 0.74, and specificity of 0.79. Additionally, the model grouped TTS patients into six clusters based on risk factors, aiding prognosis. This innovative machine-learning approach shows promise for improved TTS patient care and risk assessment.

InterTAK-ML Machine Learning Model for Predicting In-Hospital Death Risks in Takotsubo Syndrome Patients: Key Variables, Validation Results, and Prognostic Clusters

NEUROLOGY

Review explores how gut barrier dysfunction, ENS immunity and microbiome mechanisms may contribute to MS and neurologic autoimmunity.

Gut Dysfunction and Autoimmune Neurologic Diseases | MS Review

A 2026 review in Annals of Neurology examines how intestinal barrier dysfunction, the enteric nervous system and mucosal immunity may contribute to multiple sclerosis and other autoimmune neurologic diseases. The authors propose that the gut may act as both a site of immune priming and a target of autoimmune injury, linking microbiome-related mechanisms with ENS-directed autoimmunity.

ENDO-DIABETOLOGY

CALERIE analysis identifies muscle genes and pathways associated with preserving strength during caloric restriction-induced weight loss.

Muscle Genes Linked to Strength Preservation During Weight Loss

A secondary analysis of the 2-year CALERIE randomized trial identified 151 genes associated with preserving knee extensor strength during caloric restriction, independent of overall weight change. Analysis of skeletal muscle samples from 42 adults highlighted genes including HSP90AA1 and EIF3A, with pathways involving cellular proliferation, immune regulation and protein secretion.

GYNAECOLOGY - IVF

Study finds cerebral blood-flow differences mainly distinguish sFGR types, while myocardial changes remain a research phenotype.

Cerebral and Myocardial Changes in Twin sFGR | Study

A cohort study of 82 monochorionic diamniotic twin pregnancies with selective fetal growth restriction found that Gratacós type was reflected mainly by differences in cerebral blood flow and redistribution. Myocardial changes were also observed, including in larger co-twins, but did not significantly distinguish types after correction, suggesting a potential research phenotype.

UK Biobank study finds EOSS and DMO classify obesity severity differently, particularly in early or subclinical disease.

EOSS vs DMO: Comparing Obesity Classification Frameworks

A UK Biobank analysis of about 411,000 participants compared the Edmonton Obesity Staging System with the Lancet Commission Diagnostic Model for Obesity. Both identified obesity in roughly one-quarter of participants, but differed mainly in early-stage disease. The DMO captured broader mild or subclinical organ dysfunction, while EOSS emphasized more advanced obesity-related disease.

UROLOGY

Study finds ED in 93.8% of men with atherosclerotic disease, supporting its role as a marker of systemic vascular pathology.

Erectile Dysfunction Prevalence in Patients With Atherosclerosis

A single-centre study of 162 men with documented atherosclerotic disease found erectile dysfunction in 93.8%, with severe ED affecting 62.3%. Older age, particularly over 75 years, was associated with ED, while age, diabetes and longer ED duration were linked to greater severity. The findings support ED assessment as part of cardiovascular risk evaluation.

OPHTHALMOLOGY

Meta-analysis finds reproxalap improves tear production in DED, but frequent transient instillation-site irritation warrants consideration.

Reproxalap 0.25% for Dry Eye Disease: Meta-Analysis Findings

A systematic review and meta-analysis of six sources involving 2,230 adults found that reproxalap 0.25% significantly improved Schirmer test scores and tear-production response versus vehicle in dry eye disease. Instillation-site irritation was more frequent but generally mild and transient. Evidence certainty remained low to very low because of limited and heterogeneous data.

PULMONOLOGY

Review highlights gaps between clinical asthma outcomes and patient experiences, calling for more patient-informed trial endpoints.

Clinical Outcomes and Patient Experiences in Asthma: Review

A targeted review examined how asthma exacerbations, control and remission are defined and measured in guidelines and recent research. It found inconsistent definitions and limited incorporation of patient perspectives, particularly for remission. The authors highlight the need for standardized, patient-informed endpoints that better reflect patients’ experiences in asthma clinical trials.

DENTISTRY

Review explores S. mutans membrane vesicles and their potential roles in caries, oral microbial signaling and systemic disease.

S. mutans Membrane Vesicles: Implications for Caries Research

A review of research through August 2026 examines how Streptococcus mutans membrane vesicles contribute to oral microbial balance, intermicrobial signaling and pathogenic processes. These vesicles carry toxins and bioactive molecules that may influence host interactions. Understanding their biology could support future anticaries vaccines, targeted oral therapies and research into oral-systemic disease links.

AI analysis found physical inactivity was the strongest county-level predictor of obesity, with the model explaining 78% of variance.

Physical Inactivity Linked to County-Level Obesity Prevalence

A US county-level study using explainable artificial intelligence analyzed 34 variables to identify factors associated with obesity prevalence. The Light Gradient Boosting Machine model explained 78% of the variation across counties using 29 features. Physical inactivity was the dominant predictor, alongside smoking, excessive drinking, political ideology and regional culture.

Global data reveal distinct obesity trends in early and later adulthood across 196 countries from 1990 to 2022.

Global Obesity Trends in Early and Later Adulthood, 1990–2022

A global analysis of 196 countries and territories from 1990 to 2022 found that obesity prevalence increased substantially in both early adulthood (20–49 years) and later adulthood (≥50 years). Later adulthood showed larger increases in prevalence magnitude in 134 countries, while early adulthood had faster rates of increase in 168 countries, highlighting distinct age-related trends.

Omics research reveals how circadian disruption from night shifts may increase obesity risk and alter metabolic health.

Omics and Circadian Disruption: Obesity Risk in Night Shift Workers

A review of omics research shows how circadian disruption in night shift workers may promote obesity through reduced insulin sensitivity, increased adipogenesis and inflammation. Transcriptomics, proteomics, metabolomics and metagenomics have identified clock genes, hormones and microorganisms involved, highlighting opportunities for personalized strategies to protect metabolic health.

Review identifies barriers to cervical screening after sexual trauma and highlights trauma-informed approaches to improve care.

Trauma-Informed Care and Cervical Screening: Review of Barriers

A PRISMA 2020 review of 33 studies found that women with sexual trauma histories may experience greater fear, anxiety, pain, discomfort and barriers during pelvic examinations and cervical screening. Explicit consent, shared control, flexible pacing and continuity of care may improve tolerability and access, while evidence for women with sexual pain disorders remains limited.

Major weight loss or gain was associated with higher dementia risk in younger adults with diabetes and depression.

Weight Change and Dementia Risk in Young Adults With T2DM

A nationwide Korean cohort of 17,851 adults aged 20–59 with type 2 diabetes and depression found that both substantial weight loss and weight gain were associated with higher subsequent dementia risk over 10.3 years. The findings suggest weight trajectories may serve as observable risk markers, although associations with extreme weight changes require cautious interpretation.

Machine learning revealed population-specific CVD risk patterns among CKD patients in China and the United States.

Machine Learning Models Identify CVD Risk Differences in CKD Patients

A cross-national analysis of 2,166 middle-aged and older adults with CKD from US NHANES and Chinese CHARLS data found that the best-performing machine learning models differed between populations. External validation showed reduced performance across countries, while SHAP analysis identified age as the leading feature in the US cohort and hypertension in the Chinese cohort.

CRITICAL CARE

Study explores MI-E for airway clearance in ventilated children and highlights the need for safety and outcome research.

MI-E for Airway Clearance in Critically Ill Children

A UK multi-methods study explored mechanical insufflation-exsufflation (MI-E) use in invasively ventilated critically ill children. Survey and scoping review findings showed its use was most common in acutely unwell children with neuromuscular conditions, with broader applications also reported. Researchers highlighted inconsistent reporting of settings and outcomes and called for studies on safety and wider use.

Machine learning‐based pr...

Machine learning‐based prediction of in‐hospital death for patients with takotsubo syndrome