Prototyping AI-Infused Annotation Tool for Primary Ciliary Dyskinesia Diagnostics

Primary ciliary dyskinesia (PCD) is a rare genetic disorder that affects patients’ respiratory abilities. Currently, diagnosing PCD is a highly manual task that requires doctors to analyze dozens of transmission electron microscopy (TEM) images and manually gather statistics. In this paper, we propose a user interface (UI) for effective PCD diagnostics using digitized histopathology images. The repetitive and time-consuming nature of these tasks suggests, that artificial intelligence (AI) could be helpful in anomaly detection and subsequent quantification for diagnostic purposes. We developed a vertical prototype of a diagnostic tool, where we incorporated an interaction model between domain experts and AI. Development of this tool is guided by double-diamond design framework that was modified for use in the domain of medicine and to enhance the domain knowledge transfer. This paper concludes, that along with ongoing digitalization in the field of histopathology, AI-infused tools such as ours may help boost the diagnostic process of medical diseases like PCD.