Name
Abstract | ||
|---|---|---|
In this paper, we develop a computer-based solution for automatic analysis of respiratory sounds captured using the stethoscope, which has many potential applications including telemedicine and self-screening. Three types of respiratory sounds (e.g. wheezes, crackles, and normal sounds) were captured from 60 patients by a custom-built prototype device. We extracted 46 features from time, frequency and Cepstral domain from window frames and the optimal features are selected. Then a two-stage pipeline on Gaussian Mixture Model to classify these three respiratory sounds is proposed and the optimal initial parameters of GMM for each sound type are empirically calculated. By comparing with 24 FMCC features, the evaluation results show that all features proposed in this paper improved accuracy by 7.4% for the crackles and 3% for wheeze classification. On average the method for classifying wheezes, crackles and normal sounds achieved the accuracy of 98.4%, which means the models could be used in the real-world situation for the diagnosis of pulmonary diseases. |
Year | Venue | Keywords |
|---|---|---|
2017 | 2017 IEEE SMARTWORLD, UBIQUITOUS INTELLIGENCE & COMPUTING, ADVANCED & TRUSTED COMPUTED, SCALABLE COMPUTING & COMMUNICATIONS, CLOUD & BIG DATA COMPUTING, INTERNET OF PEOPLE AND SMART CITY INNOVATION (SMARTWORLD/SCALCOM/UIC/ATC/CBDCOM/IOP/SCI) | respiratory sounds classification, GMM, feature extraction, wheezes, crackles |
Field | DocType | Citations |
Stethoscope,Crackles,Respiratory sounds,Electronic stethoscope,Computer science,Speech recognition,Feature extraction,Wheeze,Hidden Markov model,Mixture model,Distributed computing | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 7 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yongpeng Liu | 1 | 2 | 1.71 |
| Yusong Lin | 2 | 0 | 1.01 |
| Xingjin Zhang | 3 | 0 | 0.68 |
| Zongmin Wang | 4 | 19 | 11.75 |
| Yang Gao | 5 | 0 | 0.34 |
| Guanling Chen | 6 | 999 | 61.99 |
| Haoyi Xiong | 7 | 505 | 44.63 |