Issue link: https://www.e-digitaleditions.com/i/1543258
40 February 2026 Inhalation Results of their work have been published in the International Journal of Medical Informatics. Reference Yu L, Hu Z, Lin Z. Rapid machine learning model for differentiating asthma and chronic obstructive pulmonary disease using age and blood parameters. International Journal of Medical Informatics. 205:106114(2026). www.science direct.com/science/article/abs/pii/ S1386505625003314?via%3Dihub. COVID-19 infection increases risk of respiratory diseases but vaccination can be protective A study of approximately one million patient records demon- strated that COVID-19 infection can increase the risk of respira- tory type-2 inflammatory dis- eases (T2IDs), including allergic asthma. However, vaccination was shown to provide protection. Led by the Karolinska Institutet of Stockholm, Sweden, the proj- ect included partners in Germany, Spain and Israel. Results were published in the Journal of Allergy and Clinical Immunology. Utilizing a global database e study used the TriNetX (Cam- bridge, Massachusetts, United States) global, federated health- care database and platform, which contained records for more than 118 million patients. From those, three cohorts were identified: indi- viduals with COVID-19 infec- tion (n = 973,794), individuals with COVID-19 vaccination (n = 691,270) and unexposed controls (n = 4,388,409). COVID-19 infection was associ- ated with a significantly increased risk for respiratory T2IDs com- pared with controls within a that would use patient age and routine blood tests to support this decision-making. Data used for analysis e team retrospectively obtained clinical and laboratory data from 9,038 patients (aged 40-98 years) who were diagnosed with COPD (6,216 patients) or asthma (2,822 patients) at the Datansha Branch of the First Affiliated Hospital of Guangzhou Medical University in 2022 to 2024. e data was split into two cohorts: training (6,326 cases, 70%) and internal valida- tion (2,712 cases. 30%). Development of the model After initial features were selected, several machine-learning models were trained and evaluated. en the best-performing option was validated to ensure its accuracy and generalizability. e model identified seven pre- dictors: patient age and six blood parameters including mean corpus- cular hemoglobin concentration (MCHC), lymphocyte percentage (LYMPH), hemoglobin content (HGB), plateletcrit (PCT), mono- cyte percentage (MONO) and eosinophil percentage (EO). In testing, the model achieved AUCs of 0.890 (training), 0.871 (internal validation), and 0.855 (external validation), with accu- racies of 0.810, 0.804 and 0.792, respectively. In addition, calibra- tion and decision curve analyses confirmed strong predictive reli- ability and clinical utility. Benefits of the model e research team has also devel- oped an online tool for rapid application. ey believe that by leveraging readily available blood test data, the model can facilitate prompt and accurate diagnosis. is can enable clinicians to begin treat- ment quickly and improve patient outcomes in acute care. avoid promoting glucocorticoid insensitivity so that medications traditionally used to treat asthma could be more effective. References Content for this article was based on, excerpted and quoted from: • Higgins ET, Busse WW, Esnault S, et al. Fueling the fire in the lung-brain axis: e salience net- work connects allergen- provoked TH17 responses to psychological stress in asthma. Brain, Behav- ior, and Immunity. 128:276-288 (2025). https://www.sciencedi rect.com/science/article/abs/pii/ S0889159125001266?via%3 Dihub. • omasy H. Researchers un- cover mechanisms linking the brain and lungs in asthma. May 2025. https://www.the-scientist . c o m / re s e a rc h e r s - u n c ove r - mechanisms- linking-the-brain- and-lungs-in-asthma-73041. A rapid, machine- learning model for differentiating asthma and COPD in emergency settings Asthma and chronic obstructive pulmonary disease (COPD) can have overlapping symptoms such as chest tightness, cough and wheeze. Consequently, differen- tiating between them based on symptoms alone can be clinically challenging, particularly in emer- gency or resource-limited settings. Spirometry is considered the "gold standard" but can be impractical in urgent care if patient cooperation and resources are limited. Still, accurate diagnosis is critical for initiating appropriate treatment. Recognizing this need, a research team from the First Affiliated Hospital of Guangzhou Medical University in China set a goal of developing a machine-learning, rapid, accessible, diagnostic model