Evaluation of the AudibleHealth Dx AI/ML-Based Dx SaMD Using FCV-SDS in the Diagnosis of COVID-19 Illness

COVID19

2019 Novel Coronavirus Disease, 2019 Novel Coronavirus Infection, 2019-nCoV Disease, 2019-nCoV Infection, COVID-19 Pandemic, COVID-19 Pandemics, COVID-19 Virus Disease, COVID-19 Virus Infection, Coronavirus Disease 2019, Coronavirus Disease-19, SARS Coronavirus 2 Infection, SARS-CoV-2 Infection, Severe Acute Respiratory Syndrome Coronavirus 2 Infection, SARS-CoV-2, Software as Medical Device, Diagnostic Software as Medical Device, SaMD, Dx SaMD, Forced Cough Vocalization, Signal Data Signature, FCV, SDS, FCV-SDS, Delta, Omicron, Artificial Intelligence, Machine Learning, AI, ML, AI/ML, Classifier, Convolutional Neural Network, CNN, Recurrent Neural Network, RNN, Oracle, Ensemble

The AudibleHealth Dx is a diagnostic software as a medical device (Dx SaMD) consisting of an ensemble of software subroutines that interacts with a proprietary database of Signal Data Signatures (SDS), using Artificial Intelligence/Machine Learning (AI/ML) to analyze forced cough vocalization signal data signatures (FCV-SDS) for diagnostic purposes. This study will evaluate the performance of the AudibleHealth Dx in comparison to a standard of care Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) test for the diagnosis of COVID-19. Bidirectional Sanger sequencing will be used to reduce the rate of false negative and false positive results. A secondary purpose of the study will be usability testing of the device for participants and providers.

The study is a prospective, multi-site, non-inferiority trial comparing the AudibleHealth Dx to Emergency Use Authorization (EUA) approved COVID-19 RT-PCR testing to demonstrate non-inferiority of the PPA and NPA when using this device to diagnose COVID-19 illness. The AudibleHealth Dx test, the Xpert Xpress SARS-CoV-2 RT-PCR (brand name) test, and bidirectional Sanger sequencing will be performed for each participant during a single encounter. Participants and staff will be blinded to AudibleHealth Dx results and the RT-PCR status at the time of testing. No one will know both results in real-time except for the Site Coordinators and unblinded statistician specifically authorized to have these results for enrollment, audit, data tracking, and data compiling purposes. Unblinding of the results will occur after the AudibleHealth Dx, RT-PCR, and variant sequencing results have been obtained. Results for the RT-PCR test will be received by the participant according to the clinical site's protocol. Variant sequencing results will be handled by each site according to their protocol. Target enrollment for this trial will be 65 COVID-19 positive cases and 247 COVID-19 negative cases, presuming a prevalence of 0.17 for a total of 312 subjects meeting all inclusion criteria.

No related papers found yet.

United States