The Use of Machine Learning in Occupational Risk Communication for Healthcare Workers – Protocol for scoping review

ABSTRACT Introduction With the development of technology, the use of machine learning (ML), a branch of computer science that aims to transform computers into decision-making agents through algorithms, has grown exponentially. This protocol arises from the need to explore the best practices for applying ML in the communication and management of occupational risks for healthcare workers. Methods and analysis This scoping review protocol 1 details a search to be conducted in the academic databases Public Medical Literature Analysis and Retrieval System Online (PUBMED), through the Virtual Health Library (BVS): Medical Literature Analysis and Retrieval System (MEDLINE), Latin American and Caribbean Literature in Health Sciences (LILACS), West Pacific (WPRIM), Nursing Database (BDENF) and Scientific Electronic Library Online (SciELO), SCOPUS, Web of Science, IEEE Xplore Digital Library and Excerpta Medica Database (EMBASE). This scoping review protocol outlines the objectives, methods, and timeline for a review that will explore and map the existing scientific evidence and knowledge on the use of machine learning (ML) in risk communication for healthcare workers. This protocol follows the PRISMA-ScR 2 and JBI guidelines 3 for conducting scoping reviews. The guiding question of the review is: How is machine learning used in risk communication for healthcare workers? The search will use PCC (Population, Concept, Context) terms and the specific descriptors defined by each database. The narrative synthesis will describe the main themes and findings of the review. The results of this scoping review will be disseminated through publication in an international peer-reviewed scientific journal. Ethics and dissemination Ethical approval is not required; data will rely on published articles. Findings will be published open access in an international peer-reviewed journal. Strengths and limitations of this study Strengths The study allows for comprehensive mapping of existing evidence on ML in occupational risk communication. The methodology follows PRISMA-ScR and JBI guidelines, ensuring transparency and replicability. The research employs a broad search strategy across multiple databases to capture relevant studies. Limitations The accuracy of ML models is dependent on the quality of the data used. The implementation of ML in healthcare requires careful evaluation of ethical, legal, and privacy issues. Registration details OSF Registries: The protocol for this review was registered in the Open Science Framework under DOI 10.17605/OSF.IO/92SK4 (available at https://osf.io/92SK4 ).