DOI: 10.1161/circ.148.suppl_1.15437 ISSN: 0009-7322

Abstract 15437: Use of Artificial Intelligence to Decrease Data Deluge From Remote Monitoring of Cardiac Implantable Electronic Devices

Danish Bawa, Rajesh Kabra, Adnan Ahmed, shanti l bansal, Rachad Ghazal, Nicholas Pham, Ilyas Colombowala, Eric Olsen, Douglas J Darden, Naga Venkata Pothineni, Rakesh Gopinathannair, Dhanunjaya R Lakkireddy
  • Physiology (medical)
  • Cardiology and Cardiovascular Medicine

Background Remote monitoring (RM) has become the standard of care for follow up of patients with cardiac implantable electronic devices (CIED). Data deluge from RM of CIEDs is one of the major challenges faced by the cardiac device clinics. This study explores the role of a novel artificial intelligence (AI) algorithm in reducing the data deluge from CIED.

Methods: The study included remote transmissions from CIEDs being monitored at 78 clinics across USA from Jan 2023 to April 2023. De-identified data was obtained from the data repository of Octagos Health. The transmissions were initially processed by AI algorithm and were either dismissed, forwarded to the device technician (DT) for review or forwarded directly to the device clinics. The main reasons for dismissals were redundant data or clinically non-actionable and non billable information. After initial triage by AI algorithm, the DT served as the second checkpoint who either dismissed or forwarded the transmissions to the device clinic with and without alerts.

Results: A total of 42,534 patients with CIEDs were included in this study. There were 19,895 (46.8%) patients with pacemakers, 8,858 (20.8%) with ILRs, 7,012 (16.5%) with ICDs, 5,028 (11.8%) with CRT-D and 1,741 (4.1%) with CRT-P. Over a period of 4 months of RM, 273,435 transmissions were received. Of these, 57,345 (21%) transmissions were dismissed by AI and 118,546 (43.4%) were dismissed by the device technician team. As such only 97,544 (35.7%) transmissions were sent to the clinicians, of which 37,627 (13.8%) had alerts while 59,917 (21.9%) were routine transmissions. For quality control, about 15% (11,324) transmissions were randomly verified by EP for accuracy and were found to be 99% (11,193/11,324) accurate.

Conclusion In our study, the use of AI algorithm significantly decreased data deluge by appropriate triaging of RM data from CIEDs. This has a potential to streamline the workflow of cardiac device clinics and improving efficiency and patient care.

More from our Archive