(New York, NY – May 19, 2026) – At this year’s American Thoracic Society (ATS) 2026 International Conference held in Orlando, the Mount Sinai Health System has unveiled a series of groundbreaking studies that distinctly push the boundaries of pulmonary and sleep medicine. As experts congregate to discuss the evolution of respiratory care, Mount Sinai’s roster of pulmonologists, data scientists, and medical researchers are highlighting innovative methodologies that transcend traditional clinical paradigms. Their research underscores pivotal advances in understanding asthma, obstructive sleep apnea (OSA), cardiovascular implications of respiratory disorders, and the integration of artificial intelligence (AI) in managing complex pulmonary diseases.
Asthma remains a prevalent respiratory condition with exacerbations that challenge both patients and healthcare providers. A Mount Sinai team delved into the prognostic power of type 2 inflammatory biomarkers—specifically blood eosinophils (BEC) and exhaled nitric oxide (FeNO)—to predict future exacerbations in asthma patients under specialty care. These biomarkers reflect nuanced immune pathways; FeNO indicates type 2 airway inflammation and T-helper cell activity, whereas BEC quantifies circulating IL-5 and eosinophil effector cells critical in inflammatory responses. By leveraging real-world clinical data alongside genomic insights from the REGAIN cohort via advanced network analysis, the study provides a refined stratification strategy, potentially revolutionizing personalized asthma management.
In the realm of sleep-disordered breathing, the intersection of obstructive sleep apnea with neurodegenerative markers hints at intricate pathophysiological mechanisms. Mount Sinai researchers examined longitudinal changes in plasma beta-amyloid 42/40 ratios among cognitively normal older adults to assess whether the burden of sleep apnea accelerates amyloid deposition, a hallmark of Alzheimer’s disease. This longitudinal investigation integrates polysomnographic metrics with biomarker trajectories, shedding light on sleep’s role in neurocognitive decline and offering a potential avenue for early interventions targeting OSA to mitigate dementia risk.
The deployment of AI in critical care has transformed patient monitoring, yet real-world model performance over time requires vigilant assessment. Mount Sinai researchers evaluated MEWS++ (Modified Early Warning Score Plus Plus), an AI-driven clinical deterioration prediction model operationalized in a tertiary care center’s step-down units. Their analysis revealed insights into temporal model stability and performance drift, emphasizing the importance of longitudinal validation frameworks to ensure AI-driven decisions remain accurate and reliable in dynamic clinical environments, preventing adverse events through timely alerts.
A provocative shift in understanding chronic respiratory diseases was highlighted in an expert symposium challenging the rigid dichotomy between asthma and chronic obstructive pulmonary disease (COPD). Mount Sinai’s Chair of Medicine, Dr. Monica Kraft, presented emerging evidence on shared inflammatory pathways, airway remodeling processes, and immune mechanisms that blur clinical distinctions between these entities. She discussed recent biologic therapies, such as dupilumab, targeting type 2 inflammation that may benefit subsets of COPD patients with overlapping pathobiology, thereby advocating for integrated treatment guidelines informed by molecular phenotyping and imaging biomarkers.
The intricate relationship between OSA and cardiovascular outcomes remains an enigma, with prior randomized controlled trials failing to demonstrate the cardiovascular benefits of continuous positive airway pressure (CPAP) therapy. Novel analyses from the Mount Sinai team re-examined data from landmark studies including the SAVE and ISAACC trials, dissecting differential risks for myocardial infarction versus stroke and identifying phenotype-specific physiologic markers such as hypoxemia burden and autonomic dysfunction. By employing machine learning to parse heterogeneity in treatment effects, their work aims to delineate patient subpopulations most likely to derive cardiovascular protection from CPAP intervention.
Further refining cardiovascular risk prediction, Mount Sinai investigators utilized the Multi-Ethnic Study of Atherosclerosis (MESA) cohort to develop parsimonious machine learning models targeting non-sleepy individuals with OSA—those frequently underdiagnosed due to minimal daytime symptoms. Their data-driven approach incorporates polysomnography-derived metrics beyond traditional Apnea-Hypopnea Index (AHI), enhancing early identification of patients at heightened cardiovascular risk and facilitating proactive clinical management strategies tailored to silent yet high-risk disease phenotypes.
On the frontier of digital health, another Mount Sinai study applied natural language processing (NLP) utilizing enhanced large language models (LLMs) fine-tuned for healthcare settings to extract clinically relevant symptomatology and cardiovascular event data from unstructured electronic health records (EHRs). This cutting-edge methodological advancement addresses the labor-intensive bottleneck of chart reviews and holds promise for scalable, real-time surveillance of disease trajectories, particularly within vulnerable populations such as those exposed to World Trade Center dust, thereby optimizing care pathways through granular longitudinal data extraction.
Indoor air quality’s impact on maternal and fetal health also garnered attention, with research tracking associations between indoor air pollutants and maternal blood pressure fluctuations throughout pregnancy. This urban prospective cohort study illuminates the cardiovascular risks imposed by environmental exposures during a critical physiological window, emphasizing the necessity for public health policies aimed at mitigating indoor pollution and safeguarding maternal and neonatal outcomes, an area often overshadowed by ambient outdoor pollution studies.
Lung cancer diagnostics are likewise undergoing a paradigm shift with the incorporation of AI-based risk stratification tools. Mount Sinai investigators validated RADLogics, a deep learning framework employing multiple neural networks to analyze nodule characteristics from sequential CT imaging, enhancing predictive accuracy for malignancy beyond single timepoint assessments. This innovation supports more informed clinical decision-making regarding lung nodule surveillance and intervention timing, potentially reducing unnecessary invasive procedures and optimizing early cancer detection pathways.
In addition to diagnostic precision, disparities in post-imaging follow-up care for incidentally detected lung nodules were rigorously analyzed. The study found that socio-economic determinants such as race and insurance status significantly influence adherence to recommended outpatient assessments and diagnostic completeness, particularly in never-smoker populations where lung cancer risk is often underestimated. These findings underscore critical health equity concerns and call for targeted interventions to bridge gaps in cancer care delivery.
Amid these themes, the integration of AI and machine learning across pulmonary and cardiovascular domains remains a unifying thread in Mount Sinai’s submissions. Their research elucidates not merely the clinical phenotypes and pathophysiologic underpinnings of complex diseases, but also innovates on implementation science by managing heterogeneity in patient populations and enhancing real-world application and generalizability of predictive algorithms, thus bridging the gap between bench and bedside.
Mount Sinai Health System, renowned as one of the largest academic medical centers in the New York metro region, continues to lead at the confluence of clinical excellence, pioneering research, and technology-driven care. Their comprehensive approach, spanning seven hospitals and hundreds of research laboratories, harnesses AI and advanced informatics not only to elucidate disease mechanisms but also to elevate patient-centered outcomes. As lung and sleep medicine evolve, Mount Sinai’s work at ATS 2026 serves as a beacon illuminating future directions in respiratory science, inclusive of precision medicine, digital health, and health equity—catalyzing a transformation in diagnosis, treatment, and prevention strategies worldwide.
Subject of Research: Pulmonary medicine, asthma, obstructive sleep apnea, cardiovascular outcomes, artificial intelligence in healthcare, lung cancer diagnostics
Article Title: Mount Sinai Unveils Pioneering Respiratory Research at ATS 2026: From Biomarker-Driven Asthma Care to AI-Powered Cardiovascular Risk Prediction in Sleep Apnea
News Publication Date: May 19, 2026
Web References:
– https://conference.thoracic.org/
– https://ats2026.d365.events/education
References: Not provided in original content
Image Credits: Not provided in original content
Keywords: Respiratory system, lungs, airway, sleep apnea, sleep disorders, asthma, air pollution, cardiovascular risk, artificial intelligence, pulmonary medicine, lung cancer, health disparities
Tags: advanced network analysis in respiratory careAI model for lung nodule risk predictionasthma exacerbation biomarkersblood eosinophils in respiratory diseaseeffects of air pollution during pregnancyexhaled nitric oxide in asthmagenomic insights in asthma researchintegration of AI in pulmonary medicineobstructive sleep apnea cardiovascular impactpulmonary and sleep medicine innovationsREGAIN cohort asthma studytype 2 inflammatory biomarkers in asthma
