Introduction: Asthma is a chronic airway inflammatory disease characterized by variable airflow limitation and intermittent symptoms. In well-controlled asthma, auscultation and spirometry often appear normal, making diagnosis challenging. Moreover, bronchial provocation tests carry a risk of inducing acute bronchoconstriction. This study aimed to develop a non-invasive, objective, and reproducible diagnostic method using machine learning-based lung sound analysis for the early detection of asthma, even during stable periods. Methods: We designed a machine learning algorithm to classify controlled asthma patients and healthy individuals using respiratory sounds recorded with a digital stethoscope. We enrolled 120 participants (60 asthmatic, 60 healthy). Controlled asthma was defined according to Global Initiative for Asthma (GINA) criteria and was supported by normal spirometry, no pathological auscultation findings, and no exacerbations in the past three months. A total of 3600 respiratory sound segments (each 3 s long) were obtained by dividing 90 s recordings from 120 participants (60 asthmatic, 60 healthy) into non-overlapping clips. The samples were analyzed using Mel-Frequency Cepstral Coefficients (MFCCs) and Tunable Q-Factor Wavelet Transform (TQWT). Significant features selected with ReliefF were used to train Quadratic Support Vector Machine (SVM) and Narrow Neural Network (NNN) models. Results: In 120 participants, pulmonary function test (PFT) results in the asthma group showed lower FEV1 (86.9 ± 5.7%) and FEV1/FVC ratios (86.1 ± 8.8%) compared to controls, but remained within normal ranges. Quadratic SVM achieved 99.86% accuracy, correctly classifying 99.44% of controls and 99.89% of asthma cases. Narrow Neural Network achieved 99.63% accuracy. Sensitivity, specificity, and F1-scores exceeded 99%. Conclusion: This machine learning-based algorithm provides accurate asthma diagnosis, even in patients with normal spirometry and clinical findings, offering a non-invasive and efficient diagnostic tool.

Tuberculosis,treatment,mortality,multidrug-resistant tuberculosis

Thank you for forwarding the external comment regarding our published article (Machine Learning-Driven Lung Sound Analysis: Novel Methodology for Asthma Diagnosis)

Ejmi Eurasian Journal of Medical Investigation

Aim: This study evaluated the prognostic value of the Haemoglobin, Albumin, Lymphocyte, and Platelet (HALP) score for six-month mortality in acute pulmonary thromboembolism (APTE) and its added value over established risk tools.Materials and Methods: In this retrospective cohort study, 556 patients with APTE confirmed by computed tomography pulmonary angiography (CTPA) were analysed between 2017 and 2024. HALP was calculated as (haemoglobin × albumin × lymphocytes) / platelets. Non-normally distributed variables are given as median (interquartile range) and compared with the Mann–Whitney U test. Independent predictors were identified by logistic regression and discrimination by ROC analysis. The incremental value of HALP over sPESI was assessed with the integrated discrimination improvement (IDI) and continuous net reclassification improvement (NRI), and internal validity by bootstrap optimism correction.Results: Mean age was 55 ± 17 years, 42.3% were male, and six-month mortality was 12.2% (n = 68). HALP was lower in non-survivors (median 0.98 vs 3.01; p < 0.001). The AUC was 0.767 (95% CI 0.713–0.822), with an optimal cut-off of 1.91 (sensitivity 79.4%, specificity 65.2%); HALP discriminated mortality better than each of its components (DeLong p < 0.05). In multivariable analysis, HALP (OR 0.66; 95% CI 0.55–0.79; p < 0.001), lactate, pulmonary arterial pressure and age were independent predictors. Adding HALP to sPESI raised the AUC from 0.612 to 0.791 (IDI 0.098; NRI 0.862), and the optimism-corrected AUC was 0.815. HALP remained independent in the malignancy subgroup (OR 0.66; p = 0.008).Conclusions: A lower HALP score is independently associated with six-month mortality in APTE and may add prognostic information to sPESI. As a simple, low-cost index it could complement guideline-based models. Given its limited specificity and the lack of external validation, these findings are hypothesis-generating pending prospective multicentre confirmation. The proposed HALP cut-off should not be applied in routine clinical practice until it has been externally validated in prospective multicentre cohorts.

Background: Hypoxia triggers pulmonary inflammation and oxidative stress, leading to endothelial dysfunction, apoptosis and lung injury. Although the flavonoid orientin exhibits potent antioxidant and anti-inflammatory properties, its pharmacokinetic behavior remains incompletely characterized. Nanoparticle-based delivery systems may enhance efficacy. This study investigated whether Orientin delivered via chitosan nanoparticles (CNP–orientin) protects against hypoxia-induced pulmonary inflammation, oxidative stress, and tissue injury in rats. Methods: A total of 48 female Sprague–Dawley rats were randomized into four normoxic and four hypoxic groups (n = 6 per group). Rats in the hypoxic groups were exposed to intermittent hypoxia (7% O₂, 8 h/day for 7 days). Hypoxic rats were allocated to experimental groups, including an untreated hypoxia group and groups receiving orientin alone, chitosan nanoparticles alone or orientin-loaded chitosan nanoparticles (CNP–orientin), while normoxic rats served as controls. Hypoxic exposure was conducted using a controlled glove box system. At the end of the protocol, blood samples were collected for serum inflammatory marker analysis, and lung tissues were harvested to assess oxidative stress parameters, hypoxia- and inflammation-related signaling pathways, apoptosis-related gene expression, and histopathological lung injury and fibrosis. Results: Intermittent hypoxia significantly increased pulmonary HIF-1α and iNOS expression, oxidative stress markers (increased malondialdehyde and decreased superoxide dismutase and glutathione), pro-inflammatory cytokines (TNF-α, IL-1β), apoptotic signaling and histopathological lung injury compared with normoxic controls (p < 0.05). Treatment with orientin or chitosan nanoparticles alone attenuated hypoxia-associated biochemical and molecular alterations compared to untreated hypoxic rats (p < 0.05). Notably, CNP–orientin treatment significantly attenuated multiple oxidative stress, inflammatory, and apoptosis-related alterations compared with untreated hypoxic rats (p < 0.05). These protective effects were observed despite the substantially lower orientin-equivalent dose in the CNP–orientin formulation compared with free orientin. NF-κB and TNF-α expression, caspase-3 levels, antioxidant markers, and Bcl-2 expression differed significantly among hypoxic groups (p < 0.05). Histopathological lung injury and Modified Ashcroft fibrosis scores were significantly lower in the CNP–orientin group than in the untreated hypoxic group (p < 0.01 and p < 0.001, respectively). Conclusions: Orientin-loaded chitosan nanoparticles effectively mitigate hypoxia-induced pulmonary inflammation, oxidative stress, apoptosis and fibrotic injury. These findings identify CNP–orientin as a promising nanotherapeutic strategy for hypoxia-associated lung diseases and support further translational investigations.