ABSTRACT
OBJECTIVE: The main goal of this research is to use distinctive features in respiratory sounds for diagnosing Chronic Obstructive Pulmonary Disease (COPD). This study develops a classification method by utilizing inverse transforms to effectively identify COPD based on unique respiratory features while comparing the classification performance of various optimal algorithms. METHOD: Respiratory sounds are divided into individual breathing cycles. In the data standardization and augmentation phase, the CycleGAN model enhances data diversity. Comprehensive analyses for these segments are then implemented using various Wavelet families and different spectral transformations representing characteristic signals. Advanced convolutional neural networks, including VGG16, ResNet50, and InceptionV3, are used for the classification task. RESULTS: The results of this study demonstrate the effectiveness of the mentioned method. Notably, the best-performing method utilizes Wavelet Bior1.3 after standardization in combination with InceptionV3, achieving a remarkable 99.75% F1-score, the gold standard for classification accuracy. CONCLUSION: Inverse transformation techniques combined with deep learning models show significant accuracy in detecting COPD disease. These findings suggest the feasibility of early COPD diagnosis through AI-powered characterization of acoustic features. MOTIVATION AND SIGNIFICANCE: The motivation behind this research stems from the urgent need for early and accurate diagnosis of Chronic Obstructive Pulmonary Disease (COPD). COPD is a respiratory disease that poses many difficulties when detected late, potentially causing severe harm to the patient's quality of life and increasing the healthcare burden. Timely identification and intervention are crucial to reduce the progression of the disease and improve patient outcomes.
ABSTRACT
Objective: To localize and distinguish between benign and malignant tumors on MRI. Method: This work proposes a high-performance method for brain tumor feature extraction using a combination of complex network and U-Net architecture. And then, the common machine-learning algorithms are used to discriminate between benign and malignant tumors. Experiments and Results. The dataset of brain MRI of a total of 230 brain tumor patients in which 77 high-grade glioma patients and 153 low-grade glioma patients were processed. The results of classifying benign and malignant tumors achieved an accuracy of 99.84%. Conclusion: The high accuracy of experiment results demonstrates that the use of the complex network and U-Net architecture can significantly improve the accuracy of brain tumor classification. This method could potentially be useful for clinicians in aiding diagnosis and treatment planning for brain tumor patients.
ABSTRACT
The continuous-wave Doppler radar measures the movement of a chest surface including of cardiac and breathing signals and the body movement. The challenges associated with extracting cardiac information in the presence of respiration and body movement have not been addressed thus far. This paper presents a novel method based on the windowed-singular spectrum analysis (WSSA) for solving this issue. The algorithm consists of two processes: signal decomposition via WSSA followed by the reconstruction of decomposed heartbeat signals through convolution. An experiment was conducted to collect chest signals in 212 people by Doppler radar. In order to confirm the effect of reducing the large noise by the proposed method, we evaluated 136 signals that were considered to contain respiration body movements from the collected signals. When comparing to the performance of a band-pass filter, the proposed analysis achieves improved beat count accuracy. The results indicate its applicability to contactless heartbeat estimation under involving respiration and body movements.
