Artificial Intelligence (AI) Enabled, Cloud-based ECG Diagnostic Solution (ZBPro) Feasibility Testing

Feasibility Testing of a Novel AI-enabled, Cloud-based ECG Diagnostic Solution to Enable Fast and Affordable Diagnosis in Long-term Continuous Ambulatory ECG Monitoring

The proposed research is to address the accessibility and affordability of technology to capture symptomatic and asymptomatic cardiac events via Long-Term Continuous Electrocardiogram Monitoring (LTCM), and to provide physicians with full access to their patients' recorded data in a timely manner. We adopt an FDA cleared single- lead OEM patch Holter made of flexible material to make long-term wearing more comfortable and more patient compliant. The patch transmits the recorded data to our cloud-based platform: ZBPro™, where our innovative technologies reside. Our proprietary AI algorithms with innovative human interaction tools, which were developed under the National Science Foundation (NSF) Small Business Innovation Research (SBIR) Phase I (Award #: 2025951) Award analyze and interpret the recorded data to detect and annotate arrhythmia/cardiac events, and generates daily reports for physician review. The feasibility of our algorithms has been verified using ZBeats' proprietary ECG database based on the standard ANSI/AAMI EC57. Data collection and transmission has been verified in the office environment.

This proposed observational study will utilize a multidisciplinary collaboration of ZBeats Inc. (ZBTS), Stony Brook University (SBU), and Lankenau Medical Center (LMC). The study will enroll patients undergoing ECG monitoring. The primary outcome measure will be the ability to capture cardiac arrhythmias and events from participants. The prescribed FDA-cleared device will serve as the ground truth (GT) for our analyses. The detected arrhythmias and events from our solution will be compared with the findings from the ground truth to obtain our system's detection rate. The goal is to achieve a detection rate of >80% to be deemed successful.

The proposed research is to address the accessibility and affordability of technology to capture symptomatic and asymptomatic cardiac events via Long-Term Continuous Electrocardiogram Monitoring (LTCM), and to provide physicians with full access to their patients' recorded data in a timely manner. We adopt an FDA cleared single- lead OEM patch Holter made of flexible material to make long-term wearing more comfortable and more patient compliant. The patch transmits the recorded data to our cloud-based platform: ZBPro™, where our innovative technologies reside. Our proprietary AI algorithms with innovative human interaction tools, which were developed under the NSF SBIR Phase I (Award #: 2025951) Award analyze and interpret the recorded data to detect and annotate arrhythmia/cardiac events, and generates daily reports for physician review. The feasibility of our algorithms has been verified using ZBeats' proprietary ECG database based on the standard ANSI/AAMI EC57. Data collection and transmission has been verified in the office environment.

Specific Aim 1: Study Setup: Setup the study REDCap instance with role-based access under SBU's REDCap system. Prepare data collection forms. Deliver the study devices to clinical sites. Provide training to clinical personnel regarding the study, procedure, informed consent and data collection. Project team assess the protocols and training are in place for the study.

Specific Aim 2: Acceptability Evaluation: Identify and evaluate the device acceptability in up-to-7-day wearing. 60-75 participants will be enrolled during 6-7-month study period. Participants' enrollment, follow-up and closeout questionnaires regarding the acceptability will be conducted in-person or via telephone by the research associate in each clinical site. Success of Aim 2: the mean acceptability percentage of the close-out survey will be > 70%.

Specific Aim 3: Arrhythmia Capture Evaluation: Evaluate devices' ability to detect cardiac arrhythmia and events, as well as the ECG signal quality. Physicians' questionnaire will be completed by the physicians from each clinical site after ZBeats' reports are reviewed and compare with reports generated by the prescription device as the ground truth. The quality of signals will be assessed by an independent cardiologist. Success of Aim 3: the physician's having a 7-day acceptance rate of > 80% and a detection sensitivity of > 80%.

Specific Aim 4: Drafting the study protocol and business plan for Phase II: Conduct statistical analysis, sample size calculations, provide projections, start working on the protocol for Phase II study.

Success of this project will resolve the feasibility concern of the proposed LTCM solution in clinical practice, enabling physicians with more proactive participation in early detection of cardiac arrhythmias and accurate risk stratification. Moreover, the proposed Phase I and Phase II studies will mitigate commercialization risk by showing clinical evidence to prove its substantial equivalence to the predicate device to receive FDA 510k cleared as well as Medicare's approval.

This proposed observational study will utilize a multidisciplinary collaboration of ZBeats INC (ZBTS), Stony Brook University (SBU), and Lankenau Medical Center (LMC). The study will enroll patients undergoing ECG monitoring. Patients will be asked to wear the study monitor and be followed for up to 7 days while wearing their prescribed FDA cleared devices simultaneously. The study does not intend to use any data or information from the investigational solution to interfere, intervene or affect any clinical decisions made for the participants. The primary outcome measure will be the ability to capture cardiac arrhythmias and events from participants. The prescribed FDA-cleared device will serve as the ground truth (GT) for our analyses. The detected arrhythmias and events from our solution will be compared with the findings from the ground truth to obtain our system's detection rate. The goal is to achieve a detection rate of >80% to be deemed successful. The secondary outcome measures include: 1) assessing signal quality by conducting independent reviewer's questionnaire; and, 2) assessing user acceptability by obtaining participants' and physicians' questionnaire.

Inclusion Criteria:

1. Patients who are > 18 years of age
2. Patients who are from one of the clinical collaborators practices
3. Patients who are prescribed to undergo ECG monitoring using an FDA-cleared monitoring device
4. Patients who are post TAVR procedure, a recent history of stroke/TIA, or having cardiac related symptoms
5. Patients who are comfortable using a smartphone or have someone in the home to help with data transmission
6. Patients who have manual dexterity to be able to recharge the phone battery or someone in the home to help them
7. Patients who are English speaking

Exclusion Criteria:

1. Patients who are <18 years of age
2. Patients who are not being prescribed with cardiac monitor testing
3. Patients who refuse to sign informed consent
4. Patients who are unable to provide informed consent
5. Patients who have a pacemaker implanted
6. Patients who are presenting with any dermatitis or infected skin over left anterior thorax
7. Patients who have a history of reaction to a prior cardiac monitor device
8. Patients who are uncomfortable having a cell phone at home for the duration of study participation
9. Patients who non-English speaking