Joanne Osuchukwu’s research while affiliated with University of Cincinnati and other places

Background: The integration of artificial intelligence (AI) in health care settings demands a nuanced approach that considers both technical performance and sociotechnical factors. Objective: This study aimed to develop a checklist that addresses the sociotechnical aspects of AI deployment in health care and provides a structured, holistic guide for teams involved in the life cycle of AI systems. Methods: A literature synthesis identified 20 relevant studies, forming the foundation for the Clinical AI Sociotechnical Framework checklist. A modified Delphi study was then conducted with 35 global health care professionals. Participants assessed the checklist's relevance across 4 stages: "Planning," "Design," "Development," and "Proposed Implementation." A consensus threshold of 80% was established for each item. IQRs and Cronbach α were calculated to assess agreement and reliability. Results: The initial checklist had 45 questions. Following participant feedback, the checklist was refined to 34 items, and a final round saw 100% consensus on all items (mean score >0.8, IQR 0). Based on the outcome of the Delphi study, a final checklist was outlined, with 1 more question added to make 35 questions in total. Conclusions: The Clinical AI Sociotechnical Framework checklist provides a comprehensive, structured approach to developing and implementing AI in clinical settings, addressing technical and social factors critical for adoption and success. This checklist is a practical tool that aligns AI development with real-world clinical needs, aiming to enhance patient outcomes and integrate smoothly into health care workflows.

The integration of Artificial Intelligence (AI) in healthcare settings de-mands a nuanced approach that considers both technical performance and soci-otechnical factors. Recognizing this, our study introduces the Clinical AI Soci-otechnical Framework (CASoF), developed through literature synthesis, and re-fined via a Modified Delphi study involving global healthcare professionals. Our research identifies a critical gap in existing frameworks, which largely focus on either technical specifications or trial outcomes, neglecting the comprehensive sociotechnical dynamics essential for successful AI deployment in clinical envi-ronments. CASoF addresses this gap by providing a structured checklist that guides the planning, design, development, and implementation stages of AI sys-tems in healthcare. The checklist emphasizes the importance of considering the value proposition, data integrity, human-AI interaction, technical architecture, organizational culture, and ongoing support and monitoring, ensuring that AI tools are not only technologically sound but also practically viable and socially adaptable within clinical settings. Our findings suggest that the successful inte-gration of AI in healthcare depends on a balanced focus on both technological advancements and the socio-technical environment of clinical settings. CASoF represents a step forward in bridging this divide, offering a holistic approach to AI deployment that is mindful of the complexities of healthcare systems. The checklist aims to facilitate the development of AI tools that are effective, user-friendly, and seamlessly integrated into clinical workflows, ultimately enhancing patient care and healthcare outcomes.

Africa has a very poor doctor-to-patient ratio. At very busy clinics, doctors could see 30+ patients per day—a heavy patient burden compared with developed countries—but productivity tools such as clinical automatic speech recognition (ASR) are lacking for these overworked clinicians. However, clinical ASR is mature, even ubiquitous, in developed nations, and clinician-reported performance of commercial clinical ASR systems is generally satisfactory. Furthermore, the recent performance of general domain ASR is approaching human accuracy. However, several gaps exist. Several publications have highlighted racial bias with speech-to-text algorithms and performance on minority accents lags significantly. To our knowledge, there is no publicly available research or benchmark on accented African clinical ASR, and speech data is non-existent for the majority of African accents. We release AfriSpeech, 200hrs of Pan-African English speech, 67,577 clips from 2,463 unique speakers across 120 indigenous accents from 13 countries for clinical and general domain ASR, a benchmark test set, with publicly available pre-trained models with SOTA performance on the AfriSpeech benchmark.

The agenda for Universal Health Coverage has driven the exploration of various innovative approaches to expanding health services to the general population. As more African countries have adopted digital health tools as part of the strategic approach to expanding health services, there is a need for defining a standard framework for implementation across board. Therefore, there is a need to review and employ an evidence-based approach to inform managing challenges, adopting best approaches, and implement informed recommendations. We reviewed a variety of digital health tools applied to different health conditions in primary care settings and highlighted the challenges faced, approaches that worked and relevant recommendations. These include limited coverage and network connectivity, lack of technological competence, lack of power supply, limited mobile phone usage and application design challenges. Despite these challenges, this review suggests that mHealth solutions could attain effective usage when healthcare workers receive adequate onsite training, deploying applications designed in an intuitive and easy to understand approach in a manner that fits into the users existing workflows, and involvement of the stakeholders at all levels in the design, planning, and implementation stages of the interventions.