A Pilot Study for the Collection Of Vocalized Individual Digital Cough Sounds from patients with suspected COVID-19 in India
试验速览
- 阶段
- 不适用
- 状态
- 进行中（未招募）
- 入组人数
- 8,000
- 试验地点
- 45
- 主要终点
- Collection of cough sound recordings, current medical symptoms, and medical history on a single occasion from 120 COVID-19 negative or positive participants as identified by PCR.
研究概览
简要总结
The novel coronavirus (SARS-CoV-2) has spread rapidly around the globe and caused widescale physical, mental, and economic damage. A key component for countries to effectively navigate this pandemic is the ability to screen for COVID-19 disease and triage individuals en masse rapidly and effectively. This ability to rapidly screen for COVID-19 is beginning to emerge with widespread testing capacity, however cost and access is a major rate limiting factor – particularly for developing world countries – and ultimately these tests can’t predict disease severity.
The technology currently exists to effectively screen for diseases such as pneumonia, asthma, and COPD using a combination of unique sound patterns contained within cough sounds and subject reported symptoms, and thus we believe the same machine learning technology can offer similar efficacy for COVID-19. If so, this screening tool would provide a rapid, safe, and low-cost mechanism to identify COVID-19 positive individuals.
This study therefore seeks to gather cough sound samples from eligible study participants and using rt-PCR and rt-qPCR as our reference standards, analyze their cough sound recordings along with subject reported symptoms to explore four hypotheses:
- There is a unique sound pattern observable in the cough sounds of COVID-19 positive individuals that is distinguishable from COVID-19 negative individuals and cough sounds of COVID-19 positive individuals change in a predictable manner over time
- An artificial intelligence algorithm with a high degree of sensitivity and specificity can be developed to indicate the presence of COVID-19 using cough sounds alone, or a combination of cough sounds and subject reported symptoms
- An algorithm with a high degree of sensitivity and specificity can be developed to triage COVID-19 positive individuals based on their likely need for medical treatment using longitudinal cough sounds alone, or a combination of longitudinal cough sounds and subject reported symptoms
- An algorithm with a high degree of sensitivity and specificity can be developed to predict the COVID-19 rt-qPCR Cycle Threshold value for COVID-19 positive individuals
研究设计
- 研究类型
- Observational
入排标准
- 年龄范围
- 18.00 Year(s) 至 99.00 Year(s)（—）
- 性别
- All
入选标准
- •be aged 18 years and older
- •be able to provide informed consent
- •be willing to follow study procedures
- •be able to provide at least 5 coughs (voluntary and/or spontaneous) and
- •Is either • an in-patient at a study site who are experiencing mild to moderate symptoms of COVID- 19 based on the ICMR guidelines, and has undergone a positive COVID19 rtPCR or rtqPCR test in the preceding 48 hours (not applicable to v 0.2 protocol) or • at a study site and is needing a COVID19 rtPCR or rtqPCR test.
排除标准
- •Participant has one or more medical contraindication to voluntary cough, including the following o Severe respiratory distress o History of pneumothorax; o Eye, chest, or abdominal surgery within 3 months of enrolling for the study o Hemoptysis (coughing up of blood) within 1 month of enrolling for the study or o Patient is requiring continuous oxygen or ventilator support.
结局指标
主要结局
Collection of cough sound recordings, current medical symptoms, and medical history on a single occasion from 120 COVID-19 negative or positive participants as identified by PCR.
时间窗: Screening/Baseline Visit, Day 2 and Day 4.
Collection of cough sound recordings, current medical symptoms, medical history, and medical treatment information on 3 distinct occasions (day 0, day 2, day 4) from 100 individuals with a known positive COVID-19 PCR result.
时间窗: Screening/Baseline Visit, Day 2 and Day 4.
To develop a post data collection algorithm to detect COVID-19 and the severity of COVID-19 and determine the accuracy of the developed algorithm with a combination of collected cough sound analysis and medical symptoms to detect COVID-19 and the severity of COVID-19 using PCR as a reference standard.
时间窗: Screening/Baseline Visit, Day 2 and Day 4.
次要结局
- N/A(N/A)