Year: 2021
Type: article
Abstract: Partially due to the pandemic, artificial Intelligence (AI) chatbots in telemedicine are a recent advancement in the medical field and are pushing the medical community forward toward automated healthcare.The purpose of this research was to discover whether AI chatbots are effective in giving a patient an idea of what medical condition he/she might have before consulting a medical professional.The hypothesis was that of the four AI chatbots that were used--Symptomate, Ada, Isabel Symptom Checker, and K Health--K Health was the one that would be the most accurate.This prediction was made based on the user interface and the accessibility of the app.Chatbots were tested by first developing a set of medical symptoms after consulting with a medical professional.The predetermined medical symptoms were input into the AI chatbots which then gave a diagnosis.Out of thirty trials, the accuracy of the chatbots were as follows.Symptomate averaged 66% correct diagnoses, Isabel Symptom Checker averaged 86% correct diagnosis, K health and ADA scored with a mean of 80% accuracy.Since the F u was less than F the null hypothesis failed to be rejected.
Author: Robert K. Swick
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W3196402214
