Home > Cardiology > HFA 2023 > AI-based technology successful in SARS-CoV-2 symptoms screening

AI-based technology successful in SARS-CoV-2 symptoms screening

Presented By
Prof. Abhinav Sharma, McGill University Health CentreMontreal, Canada
HFA 2023

The language-based AI model Alexa performed similarly to a healthcare professional in SARS-CoV-2 symptoms screening in patients with heart failure (HF) and their caregivers in the VOICE-COVID-II trial. AI-based assistance could relieve healthcare systems, at least in simple tasks like symptom screening.

COVID-19 has a significant impact on morbidity and mortality, especially among the elderly and those with HF. Moreover, the pandemic has shown that it strains healthcare resources. Can telemedicine and AI-based systems be used to alleviate pressure on healthcare systems? To answer this important question Prof. Abhinav Sharma (McGill University Health CentreMontreal, Canada) performed the VOICE-COVID-II trial (NCT04508972) to compare the accuracy of voice-based data capture by Alexa with human healthcare professionals for SARS-CoV2 symptom screening in patients with HF and their caregivers [1]. A previous study has already shown the promise of this system in healthcare-based applications [2].

In this single-centre, open-label, crossover, randomised controlled trial, 52 patients with HF or caregivers were enrolled and randomised (1:1) to either a healthcare professional or the Alexa system. This was followed by a cross-over, so patients previously assigned to the healthcare professional were assigned to Alexa and vice versa. Both the AI-based technology and the healthcare professional performed a 5-point questionnaire including a question regarding the preferred language, the purpose of the visit, and symptoms of SARS-CoV2. The primary outcome was the overall concordance level for the 5-point screening questionnaire between the AI system and the research coordinator.

Baseline characteristics revealed a male preponderance with a median age of 51 years; 40% of the participants were patients with HF, and the remaining participants were caregivers. Baseline characteristics were evenly distributed between groups.

In total, 520 questions were delivered with an overall agreement of 97.7%. This agreement was similar when patients were first contacted by the coordinator and then switched to Alexa or vice versa (coordinator first 98.5% vs 96.9% Alexa first). In addition, an unweighted kappa score of 0.93 (95% CI 0.88–0.99) suggested highly correlated results.

An additionally performed post-screening survey showed that most patients and caregivers (83%) felt comfortable with the procedure and found it is easy to use. However, 25% of participants had privacy concerns regarding AI technology.

As Prof. Sharma pointed out, it is essential to demonstrate the ability of language-based AI models to complete simple tasks before moving on to more complex ones. The VOICE-COVID-II study demonstrated that the voice-based AI system performed similarly to a healthcare professional in SARS-CoV-2 symptom screening. “Our study is one of the first randomised trials of voice-based AI technologies among patients with HF and their caregivers who are often not considered in digital studies,” Prof. Sharma concluded.

    1. Sharma A, et al. Amazon Alexa to screen for SARSCoV2 symptoms. Session Late breaking clinical trials: acute heart failure and patients monitoring, Heart Failure 2023, 20–23 May, Prague, Czechia.
    2. Sharma A, et al. Eur H J Dig Health 2021;2:521–7.


Copyright ©2023 Medicom Medical Publishers 

Posted on