Lessons for Approaching Implementation of AI Systems in Clinical Settings

• Published in: 2024 IEEE 12th International Conference on Healthcare Informatics (ICHI) pp. 472 - 477
• Main Authors: Owoyemi, Ayomide, Okpara, Ebere, Okor, Milicent, Awais, Farooq, Harwani, Sagar, Salwei, Megan, Boyd, Andrew
• Format: Conference Proceeding
• Language: English
• Published: IEEE 03-06-2024
• Subjects: Clinical settings; COVID-19; Documentation; HCI; Implementation; Medical services; Pandemics; Quality assurance; Sociotechnical systems; Stakeholders; Technological innovation
• ISSN: 2575-2634
• DOI: 10.1109/ICHI61247.2024.00067

This study underscores the complexities of implementing AI in clinical care, advocating for a holistic approach that combines technical and sociotechnical elements. It stresses aligning AI tool values with health care provider and patient needs, ensuring enhancement of patient care, compliance with standards, and seamless workflow integration. Challenges like workflow mismatches and external factors like the COVID-19 pandemic highlight the need for comprehensive testing, quality assurance, and detailed vendor documentation. These are vital for understanding an AI system's capabilities and limitations, ensuring its effective use in clinical settings. The research emphasizes human-centered design in AI development, prioritizing stakeholder engagement and user-friendly interfaces. This approach facilitates smoother healthcare provider interactions with the AI system, ensuring it meets clinical needs, enhances patient care, and supports providers in a dynamic environment.