Collecting Respiratory Sound Samples From Corona Patients to Extend the Diagnostic Capability of VOQX Electronic Stethoscope to Diagnose COVID-19 Patients

Overview

Technological developments in the recent decades has enabled the integration of electronic and digital components in the stethoscope design, in an attempt to improve auditory performance and, moreover, to assist in improving user's diagnostic accuracy by incorporating computerized, digital technologies, artificial intelligence capabilities and deep-learning-based algorithms enhancing these devices. We believe that these technologies can be used to significantly improve the diagnostic performance in the primary care phase, by means of a sophisticated stethoscope that enables auscultation to sounds and signals typically found in the sub-sound frequency level. Their transformation into the sound range, and the use of artificial intelligence and machine learning techniques to characterize sound patterns that correspond to specific problems or diseases can substantially enhance the physician's or other care giver's performance to the benefit of the patients. At this stage, the software in development does not purport to make diagnostic decisions, but only to provide information that will enhance decision and diagnosis making process, therefore enable a more accurate and definitive diagnostic decision and perhaps decrease the number of additional diagnostic tests requested.

Up to 200 patients will participate in an open, prospective and multi-center study. Patients diagnosed as positive to COVID-19 will be referred to a VOQX examination. All patients will receive detailed explanation about the purpose of the examination, its impact and will provide their consent prior to the examination. The VOQX device output will have no influence on the decision-making process of the physicians and care givers. The VOQX Stethoscope membrane will be put on the patient's chest area in predefined anterior and posterior points. The data collected in the form of breath sound signals in particular infra-sound will be transferred to an external computer and processed by machine learning algorithm developed by the company. The algorithm will seek patterns typical for the diagnosed disease for each corresponding case diagnosed.