Portinari: A Data Exploration Tool to Personalize Cervical Cancer Screening

Abstract

Socio-technical systems play an important role inpublic health screening programs to prevent cancer. Cervicalcancer incidence has significantly decreased in countries thatdeveloped systems for organized screening engaging medicalpractitioners, laboratories and patients. The system automaticallyidentifies individuals at risk of developing the disease and invitesthem for a screening exam or a follow-up exam conducted bymedical professionals. A triage algorithm in the system aims toreduce unnecessary screening exams for individuals at low-riskwhile detecting and treating individuals at high-risk. Despite thegeneral success of screening, the triage algorithm is a one-sizefitsall approach that is not personalized to a patient. This caneasily be observed in historical data from screening exams. Oftenpatients rely on personal factors to determine that they are eitherat high risk or not at risk at all and take action at their owndiscretion. Can exploring patient trajectories help hypothesizepersonal factors leading to their decisions? We present Portinari, a data exploration tool to query and visualize future trajectoriesof patients who have undergone a specific sequence of screeningexams. The web-based tool contains (a) a visual query interface(b) a backend graph database of events in patients' lives (c)trajectory visualization using sankey diagrams. We use Portinarito explore diverse trajectories of patients following the Norwegiantriage algorithm. The trajectories demonstrated variable degreesof adherence to the triage algorithm and allowed epidemiologiststo hypothesize about the possible causes.