Usability

The Usability principle states that the medical AI solutions should be usable, acceptable and deployable for the end-users in real-world practice, such as physicians, specialists, data managers and other end-users. To ensure their acceptance and adoption, the AI tools should facilitate the image analysis tasks, including the visualisation and interpretation of complex medical images, with increased productivity and satisfaction. It is important that the AI algorithms are developed while taking into account the human factors for each clinical task, by employing human-centred approaches with user engagement throughout the AI development process, including iterative usability testing. To ensure the deployability of medical AI tools, effectiveness must be estimated and integration into current clinical workflows must be demonstrated. To ensure the Usability of AI solutions in medicine and healthcare, we propose the following recommendations:

  1. User requirements: Requirements for clinical usability should be compiled with target end-users, clinical experts and relevant stakeholders. This enables understanding clinical and user needs and usability requirements, including the necessity and intended use of the AI tool, user interfaces and human-AI interactions, integration with clinical workflows, human oversight and patient requirements.
  2. User manuals: Produced for the end-users of the AI tools, AI manufacturers should develop these manuals, but end-users should be involved in the assessment of these user manuals, including material for training the end-users.
  3. Clinical evaluation: Medical AI tools should be tested with end-users of different characteristics (e.g. age, experience, sex/gender, role). It should be demonstrated that the AI tool can be used and integrated within the existing clinical workflows. Usability criteria (both qualitative and quantitative) should be defined and measured, such as number of times users agree with the AI recommendations.
  4. Clinical utility validation: Clinical utility of the AI tool should be evaluated, e.g., va comparison to standard-of-care. It should be demonstrated that the AI tool has benefits for the patient (e.g. earlier diagnosis or better outcomes) and/or clinician (e.g. increased productivity or confidence).