Detection of Obstructive Respiratory Abnormality Using Flow–Volume Spirometry and Radial Basis Function Neural Networks

Assessment of Multi-Layer Perceptron Neural Network for Pulmonary Function Test’s Diagnosis Using ATS and ERS Respiratory Standard Parameters

The aim of the research work is to investigate the operability of the entire 23 pulmonary function parameters, which are stipulated by the American Thoracic Society (ATS) and the European Respiratory Society (ERS), to design a medical decision support system capable of classifying the pulmonary function tests into normal, obstructive, restrictive, or mixed cases. The 23 respiratory parameters specified by the ATS and the ERS guidelines, obtained from the Pulmonary Function Test (PFT) device, were employed as input features to a Multi-Layer Perceptron (MLP) neural network. Thirteen possible MLP Back Propagation (BP) algorithms were assessed. Three different categories of respiratory diseases were evaluated, namely obstructive, restrictive, and mixed conditions. The framework was applied on 201 PFT examinations: 103 normal and 98 abnormal cases. The PFT decision support system’s outcomes were compared with both the clinical truth (physician decision) and the PFT built-in diagnostic software. It yielded 92–99% and 87–92% accuracies on the training and the test sets, respectively. An 88–94% area under the receiver operating characteristic curve (ROC) was recorded on the test set. The system exceeded the performance of the PFT machine by 9%. All 23 ATS\ERS standard PFT parameters can be used as inputs to design a PFT decision support system, yielding a favorable performance compared with the literature and the PFT machine’s diagnosis program.

RbfDeSolver: A Software Tool to Approximate Differential Equations Using Radial Basis Functions

A new method for solving differential equations is presented in this work. The solution of the differential equations is done by adapting an artificial neural network, RBF, to the function under study. The adaptation of the parameters of the network is done with a hybrid genetic algorithm. In addition, this text presents in detail the software developed for the above method in ANSI C++. The user can code the underlying differential equation either in C++ or in Fortran format. The method was applied to a wide range of test functions of different types and the results are presented and analyzed in detail.

A Two-Phase Evolutionary Method to Train RBF Networks

This article proposes a two-phase hybrid method to train RBF neural networks for classification and regression problems. During the first phase, a range for the critical parameters of the RBF network is estimated and in the second phase a genetic algorithm is incorporated to locate the best RBF neural network for the underlying problem. The method is compared against other training methods of RBF neural networks on a wide series of classification and regression problems from the relevant literature and the results are reported.

Deep Learning-Based Analytic Models Based on Flow-Volume Curves for Identifying Ventilatory Patterns

Introduction

Spirometry, a pulmonary function test, is being increasingly applied across healthcare tiers, particularly in primary care settings. According to the guidelines set by the American Thoracic Society (ATS) and the European Respiratory Society (ERS), identifying normal, obstructive, restrictive, and mixed ventilatory patterns requires spirometry and lung volume assessments. The aim of the present study was to explore the accuracy of deep learning-based analytic models based on flow–volume curves in identifying the ventilatory patterns. Further, the performance of the best model was compared with that of physicians working in lung function laboratories.

Methods

The gold standard for identifying ventilatory patterns was the rules of ATS/ERS guidelines. One physician chosen from each hospital evaluated the ventilatory patterns according to the international guidelines. Ten deep learning models (ResNet18, ResNet34, ResNet18_vd, ResNet34_vd, ResNet50_vd, ResNet50_vc, SE_ResNet18_vd, VGG11, VGG13, and VGG16) were developed to identify patterns from the flow–volume curves. The patterns obtained by the best-performing model were cross-checked with those obtained by the physicians.

Results

A total of 18,909 subjects were used to develop the models. The ratio of the training, validation, and test sets of the models was 7:2:1. On the test set, the best-performing model VGG13 exhibited an accuracy of 95.6%. Ninety physicians independently interpreted 100 other cases. The average accuracy achieved by the physicians was 76.9 ± 18.4% (interquartile range: 70.5–88.5%) with a moderate agreement (κ = 0.46), physicians from primary care settings achieved a lower accuracy (56.2%), while the VGG13 model accurately identified the ventilatory pattern in 92.0% of the 100 cases (P < 0.0001).

Conclusions

The VGG13 model identified ventilatory patterns with a high accuracy using the flow–volume curves without requiring any other parameter. The model can assist physicians, particularly those in primary care settings, in minimizing errors and variations in ventilatory patterns.

Quantifying ventilator unloading in CPAP ventilation

BACKGROUND

The intrinsic (muscular) patient effort driving inspiration in non-invasive ventilation modes, such as continuous positive airway pressure (CPAP) therapy, has not been identified from non-invasive data. Current CPAP settings are based on clinical judgment and assessment of symptoms of respiratory distress. Non-optimal settings, including too much positive end expiratory pressure (PEEP) can cause unintended lung injury and ventilator unloading, where patient effort drops and the CPAP device enables too much work being imposed on the injured lung. Currently, there is no non-invasive means of quantifying or identifying these effects.

METHODS

A novel model-based method of ascertaining intrinsic patient work of breathing (WOB) in CPAP is developed based on linear single compartment and 2nd order b-spline models previously used in invasive ventilation modes. Results are compared to current clinical indications, such as total Imposed WOB from the CPAP device and beak length, the latter of which is the clinical metric used to indicate alveolar overdistension. Intrinsic and Imposed WOB are compared. The hypothesis is that ventilator unloading can be assessed as a decrease in Intrinsic WOB relative to Imposed WOB, as PEEP and associated ventilator unloading rise. This hypothesis is tested using 14 subjects from a CPAP trial of several breathing rates at two PEEP levels.

RESULTS

The ratio of Intrinsic to Imposed WOB, normalised per unit tidal volume, decreased with increasing PEEP (4-7 cm H₂O), capturing the expected trend of ventilator unloading. Ventilator unloading was observed across all breathing rates. Beak length measurements showed no conclusive evidence of capturing overdistension at higher PEEP or ventilator unloading.

CONCLUSIONS

Patient Intrinsic WOB in CPAP was non-invasively quantified using model-based methods, based on pressure and flow measurements. The ratio of Intrinsic to Imposed WOB per unit tidal volume clearly and consistently showed ventilator unloading across all patients and breathing rates, with Intrinsic WOB decreasing with increasing PEEP. This trend was not observed in the current clinical metric of beak length. Non-invasively quantifying Intrinsic WOB and ventilator unloading is the critical first step to objectively optimising clinical CPAP settings, patient care, and outcomes.

Respiratory resistance and reactance in adults with sickle cell anemia: Part 2-Fractional-order modeling and a clinical decision support system for the diagnosis of respiratory disorders

BACKGROUND

A better understanding of sickle cell anemia (SCA) and improvements in drug therapy and health policy have contributed to the emergence of a large population of adults living with this disease. The mechanisms by which SCA produces adverse effects on the respiratory system of these patients are largely unknown. Fractional-order (FrOr) models have a high potential to improve pulmonary clinical science and could be useful for diagnostic purposes, offering accurate models with an improved ability to mimic nature. Part 2 of this two-part study examines the changes in respiratory mechanics in patients with SCA using the new perspective of the FrOr models. These results are compared with those obtained in traditional forced oscillation (FOT) parameters, investigated in Part 1 of the present study, complementing this first analysis.

METHODOLOGY/PRINCIPAL FINDINGS

The data consisted of three categories of subjects: controls (n = 23), patients with a normal spirometric exam (n = 21) and those presenting restriction (n = 24). The diagnostic accuracy was evaluated by investigating the area under the receiver operating characteristic curve (AUC). Initially, it was observed that biomechanical changes in SCA included increased values of fractional inertance, as well as damping and hysteresivity (p<0.001). The correlation analysis showed that FrOr parameters are associated with functional exercise capacity (R = -0.57), pulmonary diffusion (R = -0.71), respiratory muscle performance (R = 0.50), pulmonary flows (R = -0.62) and airway obstruction (R = 0.60). Fractional-order modeling showed high diagnostic accuracy in the detection of early respiratory abnormalities (AUC = 0.93), outperforming spirometry (p<0.03) and standard FOT analysis (p<0.01) used in Part 1 of this study. A combination of machine learning methods with fractional-order modeling further improved diagnostic accuracy (AUC = 0.97).

CONCLUSIONS

FrOr modeling improved our knowledge about the biomechanical abnormalities in adults with SCA. Changes in FrOr parameters are associated with functional exercise capacity decline, abnormal pulmonary mechanics and diffusion. FrOr modeling outperformed spirometric and traditional forced oscillation analyses, showing a high diagnostic accuracy in the diagnosis of early respiratory abnormalities that was further improved by an automatic clinical decision support system. This finding suggested the potential utility of this combination to help identify early respiratory changes in patients with SCA.

Respiratory resistance and reactance in adults with sickle cell anemia: Correlation with functional exercise capacity and diagnostic use

BACKGROUND

The improvement in sickle cell anemia (SCA) care resulted in the emergence of a large population of adults living with this disease. The mechanisms of lung injury in this new population are largely unknown. The forced oscillation technique (FOT) represents the current state-of-the-art in the assessment of lung function. The present work uses the FOT to improve our knowledge about the respiratory abnormalities in SCA, evaluates the associations of FOT with the functional exercise capacity and investigates the early detection of respiratory abnormalities.

METHODOLOGY/PRINCIPAL FINDINGS

Spirometric classification of restrictive abnormalities resulted in three categories: controls (n = 23), patients with a normal exam (n = 21) and presenting pulmonary restriction (n = 24). FOT analysis showed that, besides restrictive changes (reduced compliance; p<0.001), there is also an increase in respiratory resistance (p<0.001) and ventilation heterogeneity (p<0.01). FOT parameters are associated with functional exercise capacity (R = -0.38), pulmonary diffusion (R = 0.66), respiratory muscle performance (R = 0.41), pulmonary volumes (R = 0.56) and airway obstruction (R = 0.54). The diagnostic accuracy was evaluated by investigating the area under the receiver operating characteristic curve (AUC). A combination of FOT and machine learning (ML) classifiers showed adequate diagnostic accuracy in the detection of early respiratory abnormalities (AUC = 0.82).

CONCLUSIONS

In this study, the use of FOT showed that adults with SCA develop a mixed pattern of respiratory disease. Changes in FOT parameters are associated with functional exercise capacity decline, abnormal pulmonary mechanics and diffusion. FOT associated with ML methods accurately diagnosed early respiratory abnormalities. This suggested the potential utility of the FOT and ML clinical decision support systems in the identification of respiratory abnormalities in patients with SCA.

Respiration Disorders Classification With Informative Features for m-Health Applications

Respiratory disorder is a highly prevalent condition associated with many adverse health problems. As the current means of diagnosis are obtrusive and ill-suited for real-time m-health applications, we explore a convenient and low-cost automatic approach that uses wearable microelectromechanical system sensor technology. The proposed system introduces the use of motion sensors to detect the changes in the anterior-posterior diameter of the chest wall during breathing function as well as extracting the informative respiratory features to be used for breathing disorders classification. Extensive evaluations are provided on six well-known classifiers with novel feature extraction techniques to distinguish among eight different pathological breathing patterns. The effects of the number of sensors, sensor placement, as well as feature selection on the classification performance are discussed. The experimental results conducted with ten subjects show the best accuracy rates of 97.50% by support vector machine and 97.37% with decision tree bagging (DTB) with all features and after feature selection, correspondingly. Furthermore, a binary classification is proposed for distinguishing between healthy people and patients with breath problems. The different assessments of classification parameters are provided by measuring the accuracy, sensitivity, specificity, F1-score and Mathew correlation coefficient. The accuracy rates above 98% suggest superior performance of DTB in binary recognition supported by the suggested new features.

Modelling the dynamics of expiratory airflow to describe chronic obstructive pulmonary disease

Chronic obstructive pulmonary disease (COPD) is characterized by expiratory airflow limitation, but current diagnostic criteria only consider flow till the first second and are therefore strongly debated. We aimed to develop a data-based individualized model for flow decline and to explore the relationship between model parameters and COPD presence. A second-order transfer function model was chosen and the model parameters (namely the two poles and the steady state gain (SSG)) from 474 individuals were correlated with COPD presence. The capability of the model to predict disease presence was explored using 5 machine learning classifiers and tenfold cross-validation. Median (95% CI) poles in subjects without disease were 0.9868 (0.9858-0.9878) and 0.9333 (0.9256-0.9395), compared with 0.9929 (0.9925-0.9933) and 0.9082 (0.9004-0.9140) in subjects with COPD (p < 0.001 for both poles). A significant difference was also found when analysing the SSG, being lower in COPD group 3.8 (3.5-4.2) compared with 8.2 (7.8-8.7) in subjects without (p < 0.0001). A combination of all three parameters in a support vector machines corresponded with highest sensitivity of 85%, specificity of 98.1% and accuracy of 88.2% to COPD diagnosis. The forced expiration of COPD can be modelled by a second-order system which parameters identify most COPD cases. Our approach offers an additional tool in case FEV1/FVC ratio-based diagnosis is doubted.

A low-cost mobile adaptive tracking system for chronic pulmonary patients in home environment

OBJECTIVE

The main objective of this study is presenting a real-time mobile adaptive tracking system for patients diagnosed with diseases such as asthma or chronic obstructive pulmonary disease and application results at home. The main role of the system is to support and track chronic pulmonary patients in real time who are comfortable in their home environment. It is not intended to replace the doctor, regular treatment, and diagnosis.

MATERIALS AND METHODS

In this study, the Java 2 micro edition-based system is integrated with portable spirometry, smartphone, extensible markup language-based Web services, Web server, and Web pages for visualizing pulmonary function test results. The Bluetooth(®) (Bluetooth SIG, Kirkland, WA) virtual serial port protocol is used to obtain the test results from spirometry. General packet radio service, wireless local area network, or third-generation-based wireless networks are used to send the test results from a smartphone to the remote database. The system provides real-time classification of test results with the back propagation artificial neural network algorithm on a mobile smartphone. It also provides the generation of appropriate short message service-based notification and sending of all data to the Web server. In this study, the test results of 486 patients, obtained from Atatürk Chest Diseases and Thoracic Surgery Training and Research Hospital in Ankara, Turkey, are used as the training and test set in the algorithm.

RESULTS

The algorithm has 98.7% accuracy, 97.83% specificity, 97.63% sensitivity, and 0.946 correlation values. The results show that the system is cheap (900 Euros) and reliable.

CONCLUSIONS

The developed real-time system provides improvement in classification accuracy and facilitates tracking of chronic pulmonary patients.

A radial basis function neural network model for classification of epilepsy using EEG signals

Epilepsy is a disorder of cortical excitability and still an important medical problem. The correct diagnosis of a patient's epilepsy syndrome clarifies the choice of drug treatment and also allows an accurate assessment of prognosis in many cases. The aim of this study is to evaluate epileptic patients and classify epilepsy groups such as partial and primary generalized epilepsy by using Radial Basis Function Neural Network (RBFNN) and Multilayer Perceptron Neural Network (MLPNNs). Four hundred eighteen patients with epilepsy diagnoses according to International League against Epilepsy (ILAE 1981) were included in this study. The correct classification of this data was performed by two expert neurologists before they were executed by neural networks. The neural networks were trained by the parameters obtained from the EEG signals and clinic properties of the patients. Experimental results show that the predictions of both neural network models are very satisfying for learning data sets. According to test results, RBFNN (total classification accuracy = 95.2%) has classified more successfully when compared with MLPNN (total classification accuracy = 89.2%). These results indicate that RBFNN model may be used in clinical studies as a decision support tool to confirm the classification of epilepsy groups after the model is developed.