Heart Failure
Conditions
Keywords
Artificial intelligence, Speech-to-text, Patient-reported data, Digital health, Patient self-care
Brief summary
The goal of this study is to see whether this type of Artificial Intelligence (AI) Voice Assistant can reliably capture patient-reported health information to support communication between patients and their healthcare team.
Interventions
OTHERArtificial intelligence speech-to-text tool
Pilot testing of speech-to-text tool using artificial intelligence
Sponsors
Stanford University
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Adults with heart failure receiving cardiology care at Stanford University
Exclusion criteria
* Diminished decision-making capacity * Inability to understand and communicate in English sufficiently to complete study procedures * Inability to complete study procedures to the best of the participant's and investigator's knowledge * Any disorder or condition that, in the opinion of the investigator would pose a risk to participant safety or interfere with the study evaluation, procedures or completion
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Accuracy of AI Voice Assistant Compared With Expert Extraction | ~3 months | Agreement between AI-generated summaries and expert-reviewed responses (study staff extraction of participants' responses from audio recordings). The responses will be scored on a numerical scale, ranging from 0-17 points with higher points representing greater accuracy of the AI-generated summaries compared to expert-reviewed responses. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Usability of AI Voice Assistant | ~3 months | Measured using the System Usability Scale (SUS), a validated 10-item questionnaire rated on a 5-point Likert scale, with total scores ranging from 0-100 |
| Agreement between AI Voice Assistant Compared With Patient Self-Reported Written Responses | ~3 months | We will compare agreement the AI Voice Assistant summary to patient-reported responses on the same data elements, including vital signs (heart rate, blood pressure, weight) and 14 symptom and health status questions. The responses will be scored on a numerical scale, ranging from 0-17 points with higher points representing higher agreement of the AI-generated summaries compared to expert-reviewed responses. |
| Kansas City Cardiomyopathy Questionnaire Scores | ~3 months | Symptom scores based on the responses to questions obtained from AI Voice Assistant and written responses. The symptom score range from 0 to 100, with higher scores meaning less symptoms. |
| Qualitative Feedback | ~3 months | Open-ended questions inquiring about participants' likes, dislikes, and suggestions for improvements |
Countries
United States
Contacts
PRINCIPAL_INVESTIGATORAlexander Sandhu, MD, MS
Stanford University
Outcome results
None listed