Patient Perspectives on Artificial Intelligence in Radiology (PPAIR)

Guy's and St Thomas' NHS Foundation Trust

Status: Not yet enrolling

Conditions: Patients Waiting for Radiology Exams

Treatments: Other: Questionnaire

Study type: Observational

Funder types: Other

Identifiers

NCT05618860

312778

Details and patient eligibility

About

The investigators will conduct a short questionnaire with patients who are waiting for radiology exams to understand their views on the use of artificial intelligence in radiology. The questionnaire will be anonymised and entirely optional. Results will be published in peer-reviewed publications and inform future implementation of AI in clinical radiology.

Full description

The project entails a patient questionnaire. Patients will firstly be informed about the study via a member of the radiology care team. Informed consent will be obtained by a member of the team. Each participant will be assigned a unique identifier number upon recruitment. Aside from the signed consent form, no identifiable information or medical details will be collected. Consent documentation will be stored within a locked drawer in the research department of the radiology department in GSTT. The signed consent document will be kept entirely separate and will not be linked in any way to the questionnaire answers. The questionnaire data will therefore be anonymised data. A document containing the following items will be created on a GSTT computer and updated as the study progresses:

1. Participant Unique Identifier
2. Questionnaire Answers 5.2 Questionnaire When the study identifier number is assigned, it will be entered at the top of a paper questionnaire. The questionnaire is thus anonymised from the beginning of the study. The questionnaire will include demographic variables and questions with multiple choice responses corresponding to a Likert scale, which will be completed by the patient. This information will be transferred to an NHS computer in the radiology department in GSTT.

Survey data will not include identifiable information. A Gaussian Graphical Model will be inferred indicating conditional dependencies between demographic variables and participant responses. This will be performed using the desparsified Graphical LASSO method of Jankova, implemented via the R package SILGGM.

Enrollment: 175 estimated patients

Sex: All

Ages: 18+ years old

Volunteers: Accepts Healthy Volunteers

Inclusion criteria

1. Patients presenting to the radiology department waiting area for an elective appointment will be eligible to participate.
2. Participants must be aged above 18 years.
3. Participants must have independent capacity to consent to the study.

Exclusion criteria

1. Inpatients or acute patients presenting to the radiology department waiting area.
2. Those who cannot complete informed consent.