Benjamin D'SOuza
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Benjamin D'SOuza has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Anuradha Lala
Department of Population Health Science Icahn School of Medicine New York New York USA
1 shared publication - Ashwin Ravichandran
St. Vincent Heart Center of Indiana, Indianapolis, Indiana
1 shared publication - Christopher V. Chien
Knight Cardiovascular Institute, Oregon Health Sciences University, Beaverton (C.V.C.).
1 shared publication Craig Delaughter
Baylor Heart and Vascular Hospital, Fort Worth, USA
1 shared publication- Farooq H. Sheikh
MedStar Heart and Vascular Institute, Georgetown University School of Medicine, Washington, DC, USA
1 shared publication - Jared J. Herr
Sutter Health CPMC Center for Advanced Heart Failure Therapies, California Pacific Medical Center, San Francisco, California
1 shared publication
Latest Publications
- 15 February 2023
- Published by Springer Nature in Journal of Interventional Cardiac Electrophysiology
- Vol. 66 (7) , 1723-1728
A novel artificial intelligence based algorithm to reduce wearable cardioverter-defibrillator alarms
Background The wearable cardioverter defibrillator (WCD) is an important tool in mitigating sudden cardiac death (SCD). The WCD provides patient alarms for detected arrhythmias or electrical noise/artifact. Some patients experience frequent alarms for artifact. We sought to evaluate the effects of a novel artificial intelligence algorithm to reduce alarms related to electrical noise or artifact (advanced arrhythmia discrimination algorithm, AArD). Methods A retrospective review of a large commercial database of prescribed WCD. Patients prescribed the WCD during the years 2017 (discrimination algorithm, DA group) or during 2019 (advanced arrhythmia discrimination algorithm, AArD) were analyzed. A total of 96,000 patients were sampled, 48,000 in the control group (using standard direct algorithm, DA) and compared to 48,000 in the intervention AArD (4000 per group per month) for 12 months. The AArD further discriminates ECG signals based on a machine-learning algorithm utilizing intensity and frequency beyond the standard DA. Outcomes regarding alarms, arrhythmias, and safety were analyzed. Results The AArD algorithm was associated with a significant decrease in frequency of alarms over the course of WCD use; 54% of patients in the AARD versus 27% of DA had 0 alarms (P < .001). In the entire cohort, there was a 56% relative reduction in alarms with the use of AArD. Appropriate arrhythmia treatment time was not significantly different between the groups (44 s DA vs 45 s AArD [p = ns]). Conclusions A novel artificial intelligence algorithm reduces alarms without delaying appropriate therapy for VT/VF. These changes may improve compliance and quality of life in patients with a WCD.
- 15 February 2023
- Published by Springer Nature in Journal of Interventional Cardiac Electrophysiology
- Vol. 66 (7) , 1723-1728
A novel artificial intelligence based algorithm to reduce wearable cardioverter-defibrillator alarms
Background The wearable cardioverter defibrillator (WCD) is an important tool in mitigating sudden cardiac death (SCD). The WCD provides patient alarms for detected arrhythmias or electrical noise/artifact. Some patients experience frequent alarms for artifact. We sought to evaluate the effects of a novel artificial intelligence algorithm to reduce alarms related to electrical noise or artifact (advanced arrhythmia discrimination algorithm, AArD). Methods A retrospective review of a large commercial database of prescribed WCD. Patients prescribed the WCD during the years 2017 (discrimination algorithm, DA group) or during 2019 (advanced arrhythmia discrimination algorithm, AArD) were analyzed. A total of 96,000 patients were sampled, 48,000 in the control group (using standard direct algorithm, DA) and compared to 48,000 in the intervention AArD (4000 per group per month) for 12 months. The AArD further discriminates ECG signals based on a machine-learning algorithm utilizing intensity and frequency beyond the standard DA. Outcomes regarding alarms, arrhythmias, and safety were analyzed. Results The AArD algorithm was associated with a significant decrease in frequency of alarms over the course of WCD use; 54% of patients in the AARD versus 27% of DA had 0 alarms (P < .001). In the entire cohort, there was a 56% relative reduction in alarms with the use of AArD. Appropriate arrhythmia treatment time was not significantly different between the groups (44 s DA vs 45 s AArD [p = ns]). Conclusions A novel artificial intelligence algorithm reduces alarms without delaying appropriate therapy for VT/VF. These changes may improve compliance and quality of life in patients with a WCD.
- 2 February 2022
- Published by Elsevier in Journal of Cardiac Failure
- Vol. 28 (4) , 690-692
Author Response to “Referring Women for Advanced Heart Failure Therapies: Too Little, Too Late”
- 2 February 2022
- Published by Elsevier in Journal of Cardiac Failure
- Vol. 28 (4) , 690-692
Author Response to “Referring Women for Advanced Heart Failure Therapies: Too Little, Too Late”