Development and validation of a data visualization dashboard for automatic pain assessment and artificial intelligence analyses in cancer patients

Background: Pain is one of the most common and debilitating symptoms in cancer patients. Despite accurate assessment is fundamental for pain treatment, unidimensional and multidimensional subjective instruments have important limitations. This article aims to introduce a dashboard designed for multimodal data collection and visualization, which is essential for developing artificial intelligence (AI) models for automatic pain assessment (APA) in cancer pain. Methods: Functional and non-functional prerequisites were integrated. Concerning non-functional prerequisites, the dashboard was developed as a web app. For the creation of the mock-ups, the Figma web app was implemented. Shneiderman’s eight golden rules and Nielsen’s 10 heuristics were followed for interface design. Subsequently, a usability test was conducted by engaging 5 clinicians. Results: The dashboard was developed. The average success rate of the usability test was 80%. No major usability issues were identified. One user reported difficulties in task execution. Another user completed all tasks within the allotted time, except for the task of adding a new drug to the system. The feedback analysis revealed a lack of experience with computer systems. Potential solutions include introducing an initial tutorial for less experienced users and making the relevant fields more clearly visible. Conclusions: Since the use of AI-powered APA techniques is still in its infancy, further developments are needed for their widespread implementation in clinical practice. Data collection is the key step for these developments as it can act as ground truth for the training of automated systems. This user-friendly graphical interface can be utilized to design high-performance AI models for personalized pain management.