Untangling the graph based features for lung sound auscultation

Sankararaman, S.

doi:10.1016/j.bspc.2021.103215

Cited by 8 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the number of incoming edges to a node is referred to as indegree, the number of outgoing edges is called outdegree. A self-loop is considered an incoming and outgoing edge (21) .…”

Section: Methodsmentioning

confidence: 99%

Markov Chain: A Novel Tool for Electronic Ripple Analysis

Vijesh¹,

Kumar²,

Swapna³

et al. 2022

IJST

View full text Add to dashboard Cite

Objective: To delineate a novel surrogate approach to analyse the ripple component by constructing a complex network using the Markov model. Methods: Adjacency matrices (A) are constructed from the digital storage oscilloscope output signal of the Full-Wave (FWR) and Half-Wave Rectifiers (HWR) without and with a filter. The centrality measures -indegree, outdegree, in closeness, out closeness, Weighted Network Clustering Coefficient (WNCC) -are also computed for the Markov chain. Findings: With the increase of filter capacitance, more elements in the adjacency matrix become zero. Finally, only one matrix element corresponding to A 10,10 remains nonzero for FWR and A 20,20 for HWR, indicating the total rectification of the signal at 10 volts. The Markov chain analysis shows that as the ripple component decreases, the number of unconnected nodes increases and the self-loop of the last node increases. For the rectifier output without filtering, it is found that all the nodes are interconnected through edges. The greater the filtering efficiency, the greater the indegree and outdegree, and the lesser the incloseness, outcloseness and WNCC measures.

show abstract

“…When the number of incoming edges to a node is referred to as indegree, the number of outgoing edges is called outdegree. A self-loop is considered an incoming and outgoing edge (21) .…”

Section: Methodsmentioning

confidence: 99%

Markov Chain: A Novel Tool for Electronic Ripple Analysis

Vijesh¹,

Kumar²,

Swapna³

et al. 2022

IJST

View full text Add to dashboard Cite

show abstract

“…Accurate diagnosis and classification of lung diseases are crucial for proper management and treatment. The ongoing COVID-19 pandemic has added to its importance, causing both short and long-term damage to the lungs [1][2][3]. In recent years, there has been a notable rise in studies focused on understanding the pathophysiology of lung diseases, resulting in the development of diverse diagnostic and classification tools [3][4][5].…”

Section: Introductionmentioning

confidence: 99%

“…The ongoing COVID-19 pandemic has added to its importance, causing both short and long-term damage to the lungs [1][2][3]. In recent years, there has been a notable rise in studies focused on understanding the pathophysiology of lung diseases, resulting in the development of diverse diagnostic and classification tools [3][4][5]. A pulmonologist diagnoses a patient's condition through anamnesis (complaints), physical examination, and, crucially, by listening to respiratory sounds (auscultation) [6].…”

Section: Introductionmentioning

confidence: 99%

A New Shapley-Based Feature Selection Method in a Clinical Decision Support System for the Identification of Lung Diseases

Kababulut,

Gürkan Kuntalp,

Düzyel

et al. 2023

Diagnostics

View full text Add to dashboard Cite

The aim of this study is to propose a new feature selection method based on the class-based contribution of Shapley values. For this purpose, a clinical decision support system was developed to assist doctors in their diagnosis of lung diseases from lung sounds. The developed systems, which are based on the Decision Tree Algorithm (DTA), create a classification for five different cases: healthy and disease (URTI, COPD, Pneumonia, and Bronchiolitis) states. The most important reason for using a Decision Tree Classifier instead of other high-performance classifiers such as CNN and RNN is that the class contributions of Shapley values can be seen with this classifier. The systems developed consist of either a single DTA classifier or five parallel DTA classifiers each of which is optimized to make a binary classification such as healthy vs. others, COPD vs. Others, etc. Feature sets based on Power Spectral Density (PSD), Mel Frequency Cepstral Coefficients (MFCC), and statistical characteristics extracted from lung sound recordings were used in these classifications. The results indicate that employing features selected based on the class-based contribution of Shapley values, along with utilizing an ensemble (parallel) system, leads to improved classification performance compared to performances using either raw features alone or traditional use of Shapley values.

show abstract