Respiratory oscillometry and functional analyses in patients with idiopathic scoliosis

Scoliosis is a condition that affects the spine and causes chest rotation and trunk distortion. Individuals with severe deformities may experience dyspnea on exertion and develop respiratory failure. Respiratory oscillometry is a simple and non-invasive method that provides detailed information on lung mechanics. This work aims to investigate the potential of oscillometry in the evaluation of respiratory mechanics in patients with scoliosis and its association with physical performance. We analyzed 32 volunteers in the control group and 32 in the scoliosis group. The volunteers underwent traditional pulmonary function tests, oscillometry, and the 6-minute walk test (6MWT). Oscillometric analysis showed increased values of resistance at 4 Hz (R4, P<0.01), 12 Hz (R12, P<0.0001), and 20 Hz (R20, P<0.01). Similar analysis showed reductions in dynamic compliance (Cdyn, P<0.001) and ventilation homogeneity, as evaluated by resonance frequency (fr, P<0.001) and reactance area (Ax, P<0.001). Respiratory work, described by the impedance modulus, also showed increased values (Z4, P<0.01). Functional capacity was reduced in the group with scoliosis (P<0.001). A significant direct correlation was found between Cobb angle and R12, AX, and Z4 (P=0.0237, P=0.0338, and P=0.0147, respectively), and an inverse correlation was found between Cdyn and Cobb angle (P=0.0190). These results provided new information on respiratory mechanics in scoliosis and are consistent with the involved pathophysiology, suggesting that oscillometry may improve lung function tests for patients with scoliosis.

Machine learning associated with respiratory oscillometry: a computer-aided diagnosis system for the detection of respiratory abnormalities in systemic sclerosis.

INTRODUCTION: The use of machine learning (ML) methods would improve the diagnosis of respiratory changes in systemic sclerosis (SSc). This paper evaluates the performance of several ML algorithms associated with the respiratory oscillometry analysis to aid in the diagnostic of respiratory changes in SSc. We also find out the best configuration for this task. METHODS: Oscillometric and spirometric exams were performed in 82 individuals, including controls (n = 30) and patients with systemic sclerosis with normal (n = 22) and abnormal (n = 30) spirometry. Multiple instance classifiers and different supervised machine learning techniques were investigated, including k-Nearest Neighbors (KNN), Random Forests (RF), AdaBoost with decision trees (ADAB), and Extreme Gradient Boosting (XGB). RESULTS AND DISCUSSION: The first experiment of this study showed that the best oscillometric parameter (BOP) was dynamic compliance, which provided moderate accuracy (AUC = 0.77) in the scenario control group versus patients with sclerosis and normal spirometry (CGvsPSNS). In the scenario control group versus patients with sclerosis and altered spirometry (CGvsPSAS), the BOP obtained high accuracy (AUC = 0.94). In the second experiment, the ML techniques were used. In CGvsPSNS, KNN achieved the best result (AUC = 0.90), significantly improving the accuracy in comparison with the BOP (p < 0.01), while in CGvsPSAS, RF obtained the best results (AUC = 0.97), also significantly improving the diagnostic accuracy (p < 0.05). In the third, fourth, fifth, and sixth experiments, different feature selection techniques allowed us to spot the best oscillometric parameters. They resulted in a small increase in diagnostic accuracy in CGvsPSNS (respectively, 0.87, 0.86, 0.82, and 0.84), while in the CGvsPSAS, the best classifier's performance remained the same (AUC = 0.97). CONCLUSIONS: Oscillometric principles combined with machine learning algorithms provide a new method for diagnosing respiratory changes in patients with systemic sclerosis. The present study's findings provide evidence that this combination may help in the early diagnosis of respiratory changes in these patients.

Oscillation Mechanics, Integer and Fractional Respiratory Modeling in COPD: Effect of Obstruction Severity.

Purpose: This research examines the emerging role of respiratory oscillometry associated with integer (InOr) and fractional order (FrOr) respiratory models in the context of groups of patients with increasing severity. The contributions to our understanding of the respiratory abnormalities along the course of increasing COPD severity and the diagnostic use of this method were also evaluated. Patients and Methods: Forty-five individuals with no history of smoking or pulmonary diseases (control group) and 141 individuals with diagnoses of COPD were studied, being classified into 45 mild, 42 moderate, 36 severe and 18 very severe cases. Results: This study has shown initially that the course of increasing COPD severity was adequately described by the model parameters. This resulted in significant and consistent correlations among these parameters and spirometric indexes. Additionally, this evaluation enhanced our understanding of the respiratory abnormalities in different COPD stages. The diagnostic accuracy analyses provided evidence that hysteresivity, obtained from FrOr modeling, allowed a highly accurate identification in patients with mild changes [area under the receiver operator characteristic curve (AUC)= 0.902]. Similar analyses in groups of moderate and severe patients showed that peripheral resistance, derived from InOr modeling, provided the most accurate parameter (AUC=0.898 and 0.998, respectively), while in very severe patients, traditional, InOr and FrOr parameters were able to reach high diagnostic accuracy (AUC>0.9). Conclusion: InOr and FrOr modeling improved our knowledge of the respiratory abnormalities along the course of increasing COPD severity. In addition, the present study provides evidence that these models may contribute in the diagnosis of COPD. Respiratory oscillometry exams require only tidal breathing and are easy to perform. Taken together, these practical considerations and the results of the present study suggest that respiratory oscillometry associated with InOr and FrOr models may help to improve lung function tests in COPD.

Combined forced oscillation and fractional-order modeling in patients with work-related asthma: a case-control study analyzing respiratory biomechanics and diagnostic accuracy.

BACKGROUND: Fractional-order (FrOr) models have a high potential to improve pulmonary science. These models could be useful for biomechanical studies and diagnostic purposes, offering accurate models with an improved ability to describe nature. This paper evaluates the performance of the Forced Oscillation (FO) associated with integer (InOr) and FrOr models in the analysis of respiratory alterations in work-related asthma (WRA). METHODS: Sixty-two individuals were evaluated: 31 healthy and 31 with WRA with mild obstruction. Patients were analyzed pre- and post-bronchodilation. The diagnostic accuracy was evaluated using the area under the receiver operating characteristic curve (AUC). To evaluate how well do the studied models correspond to observed data, we analyzed the mean square root of the sum (MSEt) and the relative distance (Rd) of the estimated model values to the measured resistance and reactance measured values. RESULTS AND DISCUSSION: Initially, the use of InOr and FrOr models increased our understanding of the WRA physiopathology, showing increased peripheral resistance, damping, and hysteresivity. The FrOr model (AUC = 0.970) outperformed standard FO (AUC = 0.929), as well as InOr modeling (AUC = 0.838) in the diagnosis of respiratory changes, achieving high accuracy. FrOr improved the curve fitting (MSEt = 0.156 ± 0.340; Rd = 3.026 ± 1.072) in comparison with the InOr model (MSEt = 0.367 ± 0.991; Rd = 3.363 ± 1.098). Finally, we demonstrated that bronchodilator use increased dynamic compliance, as well as reduced damping and peripheral resistance. CONCLUSIONS: Taken together, these results show clear evidence of the utility of FO associated with fractional-order modeling in patients with WRA, improving our knowledge of the biomechanical abnormalities and the diagnostic accuracy in this disease.

Differential diagnosis of asthma and restrictive respiratory diseases by combining forced oscillation measurements, machine learning and neuro-fuzzy classifiers.

To design machine learning classifiers to facilitate the clinical use and increase the accuracy of the forced oscillation technique (FOT) in the differential diagnosis of patients with asthma and restrictive respiratory diseases. FOT and spirometric exams were performed in 97 individuals, including controls (n = 20), asthmatic patients (n = 38), and restrictive (n = 39) patients. The first experiment of this study showed that the best FOT parameter was the resonance frequency, providing moderate accuracy (AUC = 0.87). In the second experiment, a neuro-fuzzy classifier and different supervised machine learning techniques were investigated, including k-nearest neighbors, random forests, AdaBoost with decision trees, and support vector machines with a radial basis kernel. All classifiers achieved high accuracy (AUC ≥ 0.9) in the differentiation between patient groups. In the third and fourth experiments, the use of different feature selection techniques allowed us to achieve high accuracy with only three FOT parameters. In addition, the neuro-fuzzy classifier also provided rules to explain the classification. Neuro-fuzzy and machine learning classifiers can aid in the differential diagnosis of patients with asthma and restrictive respiratory diseases. They can assist clinicians as a support system providing accurate diagnostic options.