Anand Satyajit, Sharma Vikrant, Pourush Rajeev, Jaiswal Sandeep
Electronics and Communication Engineering, Mody University of Science and Technology, India.
Mechanical Engineering, Mody University of Science and Technology, India.
Ann Med Surg (Lond). 2022 Apr;76:103519. doi: 10.1016/j.amsu.2022.103519. Epub 2022 Apr 1.
The novel coronavirus, renamed SARS-CoV-2 and most commonly referred to as COVID-19, has infected nearly 44.83 million people in 224 countries and has been designated SARS-CoV-2. In this study, we used 'web of Science', 'Scopus' and 'goggle scholar' with the keywords of "SARS-CoV-2 detection" or "coronavirus 2019 detection" or "COVID 2019 detection" or "COVID 19 detection" "corona virus techniques for detection of COVID-19", "audio techniques for detection of COVID-19", "speech techniques for detection of COVID-19", for period of 2019-2021. Some COVID-19 instances have an impact on speech production, which suggests that researchers should look for signs of disease detection in speech utilising audio and speech recognition signals from humans to better understand the condition. It is presented in this review that an overview of human audio signals is presented using an AI (Artificial Intelligence) model to diagnose, spread awareness, and monitor COVID-19, employing bio and non-obtrusive signals that communicated human speech and non-speech audio information is presented. Development of accurate and rapid screening techniques that permit testing at a reasonable cost is critical in the current COVID-19 pandemic crisis, according to the World Health Organization. In this context, certain existing investigations have shown potential in the detection of COVID 19 diagnostic signals from relevant auditory noises, which is a promising development. According to authors, it is not a single "perfect" COVID-19 test that is required, but rather a combination of rapid and affordable tests, non-clinic pre-screening tools, and tools from a variety of supply chains and technologies that will allow us to safely return to our normal lives while we await the completion of the hassle free COVID-19 vaccination process for all ages. This review was able to gather information on biomedical signal processing in the detection of speech, coughing sounds, and breathing signals for the purpose of diagnosing and screening the COVID-19 virus.
新型冠状病毒,后被重新命名为严重急性呼吸综合征冠状病毒2(SARS-CoV-2),最常见的称呼是新冠病毒病(COVID-19),已在224个国家感染了近4483万人,并被命名为SARS-CoV-2。在本研究中,我们在2019年至2021年期间,使用“科学网”“Scopus”和“谷歌学术”,以“SARS-CoV-2检测”或“2019冠状病毒检测”或“COVID-2019检测”或“COVID-19检测”“用于检测COVID-19的冠状病毒技术”“用于检测COVID-19的音频技术”“用于检测COVID-19的语音技术”为关键词进行检索。一些COVID-19病例会对语音产生影响,这表明研究人员应利用来自人类的音频和语音识别信号,在语音中寻找疾病检测的迹象,以便更好地了解病情。本综述介绍了如何使用人工智能(AI)模型对人类音频信号进行概述,以诊断、传播认知并监测COVID-19,利用传达人类语音和非语音音频信息的生物和非侵入性信号。世界卫生组织表示,在当前的COVID-19大流行危机中,开发准确、快速且成本合理的筛查技术至关重要。在这方面,某些现有研究已显示出从相关听觉噪声中检测COVID-19诊断信号的潜力,这是一个有前景的进展。作者认为,我们需要的不是单一的“完美”COVID-19检测方法,而是快速且经济实惠的检测方法、非临床预筛查工具以及来自各种供应链和技术的工具的组合,以便在我们等待为所有年龄段完成便捷的COVID-19疫苗接种过程的同时,能够安全地恢复正常生活。本综述能够收集到关于生物医学信号处理在检测语音、咳嗽声和呼吸信号以诊断和筛查COVID-19病毒方面的信息。