THE ROLE OF VIRTUAL ASSISTANTS WITH CORONAVIRUS DIAGNOSIS AND TREATMENT

2020 
SESSION TITLE: Disaster Medicine Posters SESSION TYPE: Original Investigation Posters PRESENTED ON: October 18-21, 2020 PURPOSE: SARS-CoV-2, novel coronavirus initially detected in Wuhan, China in 2019, has been identified by World Health Organization as a cause of severe acute respiratory distress syndrome in children, adults, as well as the elderly In light of rapid person to person transmission, CDC guidelines have advised to take precautions As companies move to work-from-home, medicine has also moved significantly towards tele-medicine We analyzed the efficacy of virtual assistants, including Apple Siri, Google Assistant, Windows Cortana, and Amazon Alexa in Coronavirus and their ability to diagnose and guide patients based on relevance of symptoms METHODS: Responses from virtual assistant phone applications, Apple Siri, Google Assistant, Windows Cortana, and Amazon Alexa, were tested and analyzed to assess the accuracy of given responses, gauge the overall ability of the virtual assistants to detect the symptoms of COVID-19 and guide patients accordingly The following phrases were utilized: “I have a cough”, “I have a fever”, “I have shortness of breath”, “I feel weak”, “I have diarrhea”, “How do I know if I have Coronavirus?”, and “I think I may have COVID-19 ” Responses were categorized based on level of recognition and sophistication of response and consequently classified as- no recognition, basic recognition with response providing more information on coronavirus and symptoms, and advanced response that evaluated for further symptoms and need for medical assistance RESULTS: While Windows Cortana was able to recognize only the question with the word coronavirus at a very basic level, Google Assistant performed at a higher level and was able to identify symptoms such as cough, shortness of breath, and diarrhea as symptoms of coronavirus and provided detailed information Google Assistant additionally recognized any phrase including “Coronavirus” and “COVID-19” but still gave only information on disease Meanwhile, Apple Siri and Amazon Alexa were able to respond to 2 of the 7 questions at an advanced level at detection of words, “Coronavirus” or “COVID-19” and subsequently initiated an algorithm for the user CONCLUSIONS: While Cortana and Google Assistant functioned at a basic level, both Apple Siri and Amazon Alexa, appropriately identified the level of care needed by asking user to either stay in isolation, contact a medical doctor within 24 hours, or call 911 for emergency services based on answers provided in algorithm CLINICAL IMPLICATIONS: Although artificial intelligence can be an important tool for users in light of a pandemic, there are significant limitations that need to be addressed before we can utilize these to both recognize symptoms and variety of phrases There is also a need for these tech companies to come up with a universal system which can be utilized in a situation like this so that the population at large can benefit from these advances irrespective of the device or platform they use DISCLOSURES: No relevant relationships by Jagadish Akella, source=Web Response No relevant relationships by Vaishali Mehta, source=Web Response No relevant relationships by Rajmohan Rammohan, source=Web Response No relevant relationships by Raghavendra Sanivarapu, source=Web Response No relevant relationships by Najia Sayedy, source=Web Response
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []