Scholarlink

COVID-19, also known as 2019-nCoV, has claimed countless lives across the globe. There is no specific treatment or cure for COVID-19 at this time; however living with the illness and its consequences is unavoidable. One can quickly and effectively test for COVID to determine if they have COVID-19, hence reducing budgetary and administrative limitations on healthcare systems (Mirjalili, Meraihi, Y., Gabis S. et al. Machine learning-based research for COVID-19 detection, diagnosis, and prediction: a survey. SN Comput Sci 3, 286 (2022)).According to research, a range of signs can be used to evaluate the likelihood of infection. Non-clinical techniques such as machine learning, data mining, deep learning, and other artificial intelligence technologies are among the most promising for use outside of a clinical setting ( Dang T, Han J, Xia T, Spathis D, Bondareva E, Siegele-Brown C, Chauhan J, Grammenos A, Floto R, Cicuta P, Mascolo C Exploring Longitudinal Cough, Breath, and Voice Data for COVID-19 Progression Prediction via Sequential Deep Learning: Model Development and Validation J Med Internet Res 2022;24(6):e37004; A. Ijaz, M. Nabeel, U. Masood, T. Mahmood, M. S. Hashmi, I. Posokhova, A. Rizwan, and A. Imran, “Towards using cough for respiratory disease diagnosis by leveraging artificial intelligence: A survey,” Informatics in Medicine Unlocked, p. 100,832, 2022.; K. S. Alqudaihi, N. Aslam, I. U. Khan, A. M. Almuhaideb, S. J. Alsunaidi, N. M. A. R. Ibrahim, F. A. Alhaidari, F. S. Shaikh, Y. M. Alsenbel, D. M. Alalharith et al., “Cough sound detection and diagnosis using artificial intelligence techniques: challenges and opportunities, IEEE Access, vol. 9, pp. 102 327–102 344, 2021;). These strategies can be used to improve the diagnosis and prognosis of patients infected with the 2019-NCoV pandemic. COUGHVID, a positive and negative COVID-19 audio signal database, was employed in this study to evaluate a supervised learning approach such as random forest, decision trees, and support vector machines ( H. Xiong, S. Berkovsky, M.A. Kaafar, A. Jaffe, E. Coiera, R.V. Sharan, Reliability of crowdsourced data and patient-ˆ reported outcome measures in cough-based COVID-19 screening, Sci Rep 12 (1) (Dec. 2022); F. Manzella, G. Pagliarini, G. Sciavicco, I.E. Stan, The voice of COVID-19: breath and cough recording classification with temporal decision trees and random forests, Artif Intell Med 137 (Mar. 2023);). To establish the strength of the relationship between the dependent features, the correlation coefficients between various dependent and independent variables were examined. To accurately distinguish COVID-19 patients from healthy individuals, the study used a machine learning approach that employed Mel-frequency cepstral coefficients and a neural network (Islam et al. in Biomedical Engineering Advances 3, 2022). The research produced a positive classification detection rate of 93 percent for participants and predicted the recovery period of COVID-19 patients with an accuracy rate of 91.5 percent by combining the optimal XGBoost model with a Random Forest model. The research employed a data set of 400 cough audio signals that included both positive and negative COVID-19 cases. Recent research has demonstrated the potential of using audio data—such as voice, respiration, and cough—in the COVID-19 screening process (M. Klopper, R. Warren, M. Pahar and T. Niesler, “Covid-19 detection in cough, breath and speech using deep transfer learning and bottleneck features,” Computers in biology and medicine, vol. 141, p. 105,153, 2022.24;Islam et al. in Biomedical Engineering Advances 3, 2022; N. Ibtehaz, T. Rahman, YMS. Mekki, Y. Qibalwey, S. Mahmud, M. Ezeddin MEH. Chowdhury, S. Zughaier, SASA. Al-Maadeed, QUCoughScope: an Artificially Intelligent Mobile Application to Detect Asymptomatic COVID-19 Patients Using Cough and Breathing Sounds;). Given the available audio sample, these methods can only detect infections once, and they do not keep track of COVID-19’s disease progression. There has been little research done on using longitudinal audio data to continuously track COVID-19 progression, particularly recovery (Dang T, Han J, Xia T, Spathis D, Bondareva E, Siegele-Brown C, Chauhan J, Grammenos A, Floto R, Cicuta P, Mascolo C Exploring Longitudinal Cough, Breath, and Voice Data for COVID-19 Progression Prediction via Sequential Deep Learning: Model Development and Validation J Med Internet Res 2022;24(6):e37004). Tracking illness progression traits and recovery patterns may provide insights that result in earlier therapy initiation or modification, better resource allocation in healthcare systems, and other benefits.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Supervised Learning Approach Using Random Forest to Detect Covid-19 with a Prediction of Recovery in COVID-19 Patients Using Cough Audio Signals

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mridul Vats,
  • Pooja Sabherwal,
  • Monika Agrawal,
  • Mona Aggarwal

摘要

COVID-19, also known as 2019-nCoV, has claimed countless lives across the globe. There is no specific treatment or cure for COVID-19 at this time; however living with the illness and its consequences is unavoidable. One can quickly and effectively test for COVID to determine if they have COVID-19, hence reducing budgetary and administrative limitations on healthcare systems (Mirjalili, Meraihi, Y., Gabis S. et al. Machine learning-based research for COVID-19 detection, diagnosis, and prediction: a survey. SN Comput Sci 3, 286 (2022)).According to research, a range of signs can be used to evaluate the likelihood of infection. Non-clinical techniques such as machine learning, data mining, deep learning, and other artificial intelligence technologies are among the most promising for use outside of a clinical setting ( Dang T, Han J, Xia T, Spathis D, Bondareva E, Siegele-Brown C, Chauhan J, Grammenos A, Floto R, Cicuta P, Mascolo C Exploring Longitudinal Cough, Breath, and Voice Data for COVID-19 Progression Prediction via Sequential Deep Learning: Model Development and Validation J Med Internet Res 2022;24(6):e37004; A. Ijaz, M. Nabeel, U. Masood, T. Mahmood, M. S. Hashmi, I. Posokhova, A. Rizwan, and A. Imran, “Towards using cough for respiratory disease diagnosis by leveraging artificial intelligence: A survey,” Informatics in Medicine Unlocked, p. 100,832, 2022.; K. S. Alqudaihi, N. Aslam, I. U. Khan, A. M. Almuhaideb, S. J. Alsunaidi, N. M. A. R. Ibrahim, F. A. Alhaidari, F. S. Shaikh, Y. M. Alsenbel, D. M. Alalharith et al., “Cough sound detection and diagnosis using artificial intelligence techniques: challenges and opportunities, IEEE Access, vol. 9, pp. 102 327–102 344, 2021;). These strategies can be used to improve the diagnosis and prognosis of patients infected with the 2019-NCoV pandemic. COUGHVID, a positive and negative COVID-19 audio signal database, was employed in this study to evaluate a supervised learning approach such as random forest, decision trees, and support vector machines ( H. Xiong, S. Berkovsky, M.A. Kaafar, A. Jaffe, E. Coiera, R.V. Sharan, Reliability of crowdsourced data and patient-ˆ reported outcome measures in cough-based COVID-19 screening, Sci Rep 12 (1) (Dec. 2022); F. Manzella, G. Pagliarini, G. Sciavicco, I.E. Stan, The voice of COVID-19: breath and cough recording classification with temporal decision trees and random forests, Artif Intell Med 137 (Mar. 2023);). To establish the strength of the relationship between the dependent features, the correlation coefficients between various dependent and independent variables were examined. To accurately distinguish COVID-19 patients from healthy individuals, the study used a machine learning approach that employed Mel-frequency cepstral coefficients and a neural network (Islam et al. in Biomedical Engineering Advances 3, 2022). The research produced a positive classification detection rate of 93 percent for participants and predicted the recovery period of COVID-19 patients with an accuracy rate of 91.5 percent by combining the optimal XGBoost model with a Random Forest model. The research employed a data set of 400 cough audio signals that included both positive and negative COVID-19 cases. Recent research has demonstrated the potential of using audio data—such as voice, respiration, and cough—in the COVID-19 screening process (M. Klopper, R. Warren, M. Pahar and T. Niesler, “Covid-19 detection in cough, breath and speech using deep transfer learning and bottleneck features,” Computers in biology and medicine, vol. 141, p. 105,153, 2022.24;Islam et al. in Biomedical Engineering Advances 3, 2022; N. Ibtehaz, T. Rahman, YMS. Mekki, Y. Qibalwey, S. Mahmud, M. Ezeddin MEH. Chowdhury, S. Zughaier, SASA. Al-Maadeed, QUCoughScope: an Artificially Intelligent Mobile Application to Detect Asymptomatic COVID-19 Patients Using Cough and Breathing Sounds;). Given the available audio sample, these methods can only detect infections once, and they do not keep track of COVID-19’s disease progression. There has been little research done on using longitudinal audio data to continuously track COVID-19 progression, particularly recovery (Dang T, Han J, Xia T, Spathis D, Bondareva E, Siegele-Brown C, Chauhan J, Grammenos A, Floto R, Cicuta P, Mascolo C Exploring Longitudinal Cough, Breath, and Voice Data for COVID-19 Progression Prediction via Sequential Deep Learning: Model Development and Validation J Med Internet Res 2022;24(6):e37004). Tracking illness progression traits and recovery patterns may provide insights that result in earlier therapy initiation or modification, better resource allocation in healthcare systems, and other benefits.