Current Issue : October-December Volume : 2026 Issue Number : 4 Articles : 5 Articles
Background/Objectives: Preoperative pulmonary function tests (PFTs) contain numerous physiologic parameters, yet surgeons typically rely on forced expiratory volume in one second (FEV1) and diffusing capacity of the lung for carbon monoxide (DLCO) to assess surgical risk. This study aimed to evaluate whether artificial intelligence (AI) could utilize more PFT data to predict the occurrence of prolonged air leak (PAL) following lung resection. Methods: An optical character recognition (OCR) model was used to extract structured data from PFT reports. These data were combined with clinical and demographic features from our institutional Society of Thoracic Surgeons General Thoracic Surgery Database (STS-GTSD) between 2016 and 2023. A feature selection algorithm was used to select the most predictive features, and a neural network was trained and tested on an internal validation cohort to predict PAL. Model performance was compared to previously published models. Results: There were 410 patients undergoing lung resection who had PFTs successfully digitized by the OCR system. A total of 76 available PFT features were extracted per patient. The final AI model included 10 key input variables, including three PFTs and seven clinical variables. On validation, the model achieved a specificityThis performance exceeded most existing PAL prediction models. Conclusions: AI-driven models using structured PFT and clinical data can enhance prediction of prolonged air leak after lung resection and outperform conventional regression-based models. Further research may focus on external validation and integration into clinical workflows....
Background/Objectives: Blood-based biomarkers may improve risk stratification of indeterminate pulmonary nodules detected on low-dose computed tomography (LDCT). We evaluated the diagnostic performance and independent predictive value of an aptamer- based blood assay, AptoDetect™-Lung, in a high-risk Korean screening population. Methods: This multicenter prospective cohort study enrolled adults with Lung Imaging Reporting and Data System (Lung-RADS) category 3 or 4 pulmonary nodules identified on LDCT across ten tertiary hospitals in South Korea between June 2023 and December 2024. Analyses focused on a predefined high-risk subgroup meeting Korean screening criteria (age 54–74 years and ≥30 pack-years of smoking). Baseline serum Ap-toDetect™-Lung scores were measured. Associations with lung cancer diagnosis were assessed using univariate and multivariable logistic regression, adjusting for clinical and radiologic variables. Diagnostic performance was evaluated using receiver operating characteristic analysis. Results: Among 1084 participants with histopathologic confirmation, 319 met high-risk criteria, of whom 260 (81.5%) were diagnosed with lung cancer. In this subgroup, the AptoDetect™-Lung score was independently associated with malignancy after adjustment (adjusted odds ratio of 1.14 per unit; 95% confidence interval of 1.02–1.27; p = 0.020). Discriminative performance was higher in the high-risk subgroup than in the overall cohort (area under the curve [AUC] of 0.639 vs. 0.570; p = 0.025). Performance was higher for squamous cell carcinoma and small-cell lung cancer than for adenocarcinoma. A multivariable model incorporating biomarker score, Lung-RADS category, age, and family history achieved an AUC of 0.710. Conclusions: An aptamer-based blood biomarker may provide modest adjunctive value for risk stratification in high-risk individuals....
Introduction: Large language models (LLMs) may support clinical reasoning, yet real-world outpatient studies integrating structured clinical data with chest radiographs (CXRs) remain limited. We compared three LLMs for pulmonary differential diagnosis in routine clinical practice. Methods: In this prospective, single-center observational study, consecutive adult outpatients presenting with respiratory complaints between 06 October and 31 December 2025 were enrolled. For each case, a standardized structured clinical form and de-identified CXRs were provided to three LLMs (ChatGPT-5.2, Google Gemini 3 Flash, and Microsoft Copilot) and to three blinded pulmonologists. The primary diagnosis was assigned by the examining pulmonologist, and the reference diagnosis was defined by agreement of at least two blinded pulmonologists. Concordance and Cohen’s kappa were assessed. Results: A total of 120 patients were included. Agreement among the blinded pulmonologists was high, and agreement between the primary and reference diagnoses was excellent. Compared with the reference diagnosis, ChatGPT-5.2 and Microsoft Copilot showed higher concordance than Google Gemini 3 Flash, with both demonstrating moderate overall agreement. Concordance did not differ by age or sex. Across diagnostic categories, performance was highest for pneumonia/upper respiratory tract infection and asthma. Discussion: In this real-world pulmonology outpatient cohort, ChatGPT-5.2 and Microsoft Copilot showed better diagnostic concordance than Google Gemini 3 Flash when structured clinical data and CXRs were evaluated together. These findings support the potential role of LLMs as adjunctive decision-support tools in pulmonology, while also indicating that performance remains diagnosis-dependent and insufficient to replace expert clinical judgment....
CVDs) are closely linked through shared inflammatory and metabolic pathways. Metabolic syndrome (MetS), shared by both conditions, may represent a key mechanistic link between systemic inflammation, cardiovascular disease, and COPD outcomes. Objective: To assess the prevalence and clinical correlates of MetS and examine whether increasing MetS burden is linked to more severe symptoms and a higher comorbidity load in patients with COPD. Methods: We analyzed cross-sectional data from a multicenter COPD registry. MetS was defined by the presence of at least three components: obesity, hypertension, hyperglycemia, or dyslipidemia based on clinical and medication data. Associations between MetS burden and COPD characteristics were evaluated using multivariable models adjusted for age, sex, smoking history, exacerbation status, lung function, inhaled therapy, and study center. Results: Among 5030 patients, MetS was present in 10.4% and was more frequent in women (11.9% vs. 9.6%, p = 0.01). Patients with MetS had greater symptom burden, and more frequent signs of cyanosis, cor pulmonale, and heart failure. MetS was most common in Global Initiative for Chronic Obstructive Lung Disease stage II–III and associated with higher forced expiratory volume in one second but lower forced vital capacity, with similar exacerbation rates between the groups. Cardiovascular, sleep, renal, and connective tissue comorbidities were more prevalent in patients with MetS. A dose–response relationship was observed, with each additional metabolic syndrome component independently associated with increased odds of respiratory symptoms and cardiometabolic comorbidities (all p < 0.01). Conclusions: Our findings suggest that metabolic syndrome is present in approximately 10% of patients with COPD and is associated with greater symptom burden and a higher prevalence of cardiovascular, renal, and sleep-related comorbidities. The observed stepwise relationship supports the presence of a clinically relevant cardiometabolic profile in COPD. However, given the cross-sectional registry-based design, causal inferences cannot be made, and prospective studies are needed to confirm these associations and evaluate targeted interventions....
Background: Although single-inhaler triple therapy (SITT) improves COPD control, the specific structural and behavioral predictors of short-term clinical response following therapeutic simplification remain incompletely characterized. Methods: This prospective, multicenter observational study (N = 684) evaluated patients switching from triple therapy regimens involving multiple inhalers to SITT. A clinically meaningful response was defined as an intra-individual reduction of ≥2 points in the validated RADAR score at three months. Results: Therapeutic simplification reduced regimens requiring ≥4 inhalations/day from 46.1% to 14.3%, and poor behavioral adherence from 45.2% to 16.6%. Multivariable models identified an observed responder profile: higher baseline RADAR burden was the strongest predictor of improvement (aOR 2.00), whereas severe airflow limitation (FEV1 < 50%) attenuated the response. Exploratory mediation analysis indicated that 88.6% of the observed clinical stabilization was not explained by measured adherence changes, and may therefore also encompass unmeasured behavioral, educational or device-related factors. Patients burdened with both high complexity and poor adherence showed the highest rate of combined structural–behavioral improvement (25.0% vs. 4.7% overall). Conclusions: Switching from MITT to SITT was associated with reduced treatment complexity, improved adherence profiles, and short-term improvement in RADAR-defined clinical control. Patients with greater baseline RADAR burden and regimen complexity showed larger observed improvements....
Loading....