Artificial Intelligence for radiographers: A review of current applications and a survey among Dutch hospitals

Abstract

Purpose: Artificial Intelligence (AI) has changed radiology substantially in the last years, where the focus of attention has mainly been on the radiologist profession. However, the radiographer’s role has been largely ignored while AI also is affecting for example workflow management, treatment planning and image reconstruction. Radiographers are not prepared for changes that will come with the introduction of AI into everyday work.\n \nMaterials and Method: Firstly, a survey was conducted among Dutch radiographers to investigate what role AI currently plays in their everyday work and what needs with respect to education and training currently exist. Secondly, a project was developed consisting of three main steps, leading to online AI education (e-learnings) tailored to the needs of radiographers. The first two steps in this project consist of a systematic review of the scientific literature regarding AI applications that influence the radiography workflow, and focus groups with AI experts based on the outcomes of the systematic review to obtain better insight into which developments will lead to future changes for the everyday work of radiographers.\n\nResults: The survey questionnaire was filled in by 126 radiographers from hospitals all over the Netherlands. 56% of the respondents work in Radiology, 24% in radiotherapy, and 10% in nuclear medicine. 90% is familiar with the concept of AI, and 70% encounters some form of AI in their day-to-day work. In most cases this concerns image reconstruction (40%), image recognition (35%) and image fusion (33%), but also quite often postprocessing and automatic delineation (both 29%) and dose optimization (28%). In a few instances the AI concerns patient positioning (10%), workflow management (8%) and clinical decision support (7%). Most respondents feel a need for some form of AI education (79%), preferably in the form of an e-learning (71%). The top three educational topics are the application of AI tools (93%), the basic principles of AI (79%), and the safe use of AI algorithms (70%).\nIn order to fulfill the need for online specific radiographer-oriented education in AI a project has been funded by the Dutch Taskforce for Applied Research (SIA) that will run until March 2023. Here, preliminary results will be presented on the first step of the project: the systematic review of the scientific literature. For the systematic review a total of 70 articles were found, ranging from review, prospective, retrospective to survey articles in search engines like PubMed, Scopus and Google Scholar. Results show a wide variety of applications of AI that (will) influence the work of radiographers, ranging from changes in everyday workflow, like patient checks, planning of examinations, acquisition of images and post-processing activities, to changes in work flexibility, like cross-modality employability or performing radiologist tasks, and training, implementing and quality control of AI systems. Knowledge of AI, the basics as well as pitfalls, challenges, ethical and legal complications is prerequisite for radiographers.\n \nConclusions: A survey among Dutch radiographers shows that they often encounter AI applications in their everyday work. They indicate a need for (preferably online) education to increase their knowledge about AI. A project has been funded to fulfil this wish. The first step of this project (a systematic review) has been taken and it should eventually lead to radiographer-specific e-learnings in 2023.

Ähnliche Arbeiten