Exploring Stakeholder Requirements to Enable the Research and Development of Artificial Intelligence Algorithms in a Hospital-Based Generic Infrastructure: Protocol for a Multistep Mixed Methods Study

Background In recent years, research and developments in advancing artificial intelligence (AI) in health care and medicine have increased. High expectations surround the use of AI technologies, such as improvements for diagnosis and increases in the quality of care with reductions in health care costs. The successful development and testing of new AI algorithms require large amounts of high-quality data. Academic hospitals could provide the data needed for AI development, but granting legal, controlled, and regulated access to these data for developers and researchers is difficult. Therefore, the German Federal Ministry of Health supports the Protected Artificial Intelligence Innovation Environment for Patient-Oriented Digital Health Solutions for Developing, Testing, and Evidence-Based Evaluation of Clinical Value (pAItient) project, aiming to install the AI Innovation Environment at the Heidelberg University Hospital in Germany. The AI Innovation Environment was designed as a proof-of-concept extension of the already existing Medical Data Integration Center. It will establish a process to support every step of developing and testing AI-based technologies. Objective The first part of the pAItient project, as presented in this research protocol, aims to explore stakeholders’ requirements for developing AI in partnership with an academic hospital and granting AI experts access to anonymized personal health data. Methods We planned a multistep mixed methods approach. In the first step, researchers and employees from stakeholder organizations were invited to participate in semistructured interviews. In the following step, questionnaires were developed based on the participants’ answers and distributed among the stakeholders’ organizations to quantify qualitative findings and discover important aspects that were not mentioned by the interviewees. The questionnaires will be analyzed descriptively. In addition, patients and physicians were interviewed as well. No survey questionnaires were developed for this second group of participants. The study was approved by the Ethics Committee of the Heidelberg University Hospital (approval number: S-241/2021). Results Data collection concluded in summer 2022. Data analysis is planned to start in fall 2022. We plan to publish the results in winter 2022 to 2023. Conclusions The results of our study will help in shaping the AI Innovation Environment at our academic hospital according to stakeholder requirements. With this approach, in turn, we aim to shape an AI environment that is effective and is deemed acceptable by all parties. International Registered Report Identifier (IRRID) DERR1-10.2196/42208


Current projects
• Could you please describe very briefly which projects dealing with the topics of AI and health care data you are currently working on (or planning)? • Which hardware and software do you currently use for AI/ML? • What are your or your organization's goals in these projects? (e.g., research, selling patents or licenses, etc…)

Data usage
• To what extent do you need data from the Heidelberg University Hospital or comparable institutions for your projects? • What kind of data do you need? • How could data from external sources support your projects?
• What are possible advantages of data from an academic hospital? In this context, how relevant is the data storage location, for example if data are stored in the hospital's or your infrastructure? • What difficulties have you encountered in the past in the context of these projects?
• To what extent have you encountered barriers in working with these kind of data? • Concerning intellectual property, what role does the data provider play?

Data provider
• What kind of support from the data provider would help you in conducting your projects? • What are necessary requirements that have to be met in order for you to work within the infrastructure and framework conditions of the AI Innovation Environment?
Weinert L, Klass M, Schneider G, Heinze O. Exploring Stakeholder Requirements to enable research and development of AI algorithms in a hospital based generic in-frastructure: Research Protocol for a multi-step mixed-methods study

Conclusion
Do you have any further recommendations for the AI Innovation Environment?
Are there any other thoughts you would like to share with us at this time?
Thank you so much for the conversation and your interesting opinions!

Introduction:
Today, I would like to talk with you about the topic of "artificial intelligence" and data usage for the development of "artificial intelligence". What are your first associations with these topics?
Now, I would like to read you a short definition of artificial intelligence: "Artificial intelligence", also called AI, is part of the subject area of informatics. AI means, that a computer is able to fulfill tasks under supervision and learn from data and observations, which would normally require human intelligence. They can adapt autonomously when being "fed" big amounts of data. On this basis and when confronted with specific questions, they are able to detect patterns and make predictions or decisions.
Through AI, a computer could perform the following tasks: suggest new songs based on known musical preferences process billings in a hospital administration in an automized manner compare moles with a big data base during skin cancer screening and suspect skin cancer Do you have any questions regarding this explanation and the examples?

Scenario: AI Innovation Environment
Imagine you are a new patient at our hospital and came in for an appointment. Here you are informed about a new infrastructure for the development of AI technologies that is being tested at the hospital, in corporation with companies and other research institutions. Before your treatment, you will be asked if your data can be used for the development of AI.
What are your initial thoughts?
In general, which data would you release for the development of AI? Which data would you prefer not to release?
Would you like to be informed about the use of your data? Should your approval be collected beforehand?
What could lead to your refusal of releasing data? What could lead to your approval? Weinert L, Klass M, Schneider G, Heinze O. Exploring Stakeholder Requirements to enable research and development of AI algorithms in a hospital based generic in-frastructure: Research Protocol for a multi-step mixed-methods study To what extent would these factors be different if the data were not supposed to be used to for AI development, but for "usual" software?
Would your answers be different if external companies are involved? Or if only publicly funded research institutions are involved?

Reflection
In the past 30 minutes, we talked about AI in medicine. What, if anything, have you learned from this conversation? Has something changed in your view on the topic of AI in medicine or data usage?

Conclusion
Do you have any further recommendations or thoughts on this topic?
Are there any other thoughts you would like to share with us at this time?
Thank you so much for the conversation and your interesting opinions!