Abstract:

Advancement in algorithms using artificial intelligence (AI) has driven the effective application of machine learning methods to forecast, diagnose, and treat diseases, accelerating digital transformation in healthcare. Early detection is becoming increasingly crucial in oncology, because it offers "enormous potential to increase patient survival and decrease overall mortality" (Placido et al., 2023). For this reason, the case of follow-up of pediatric cancer patients after treatment to identify early relapses is a constant challenge. Traditional methods, such as population-based screening, are often "impractical" because of "costly clinical examinations for a large number of patients with false positive prognoses" (Placido et al., 2023). In addition, AI overcomes regional barriers with limited access to specialized centers by providing solutions that overcome "financial and technological" barriers (Xue et al., 2023).

The ML algorithm has shown its ability to predict cancer risk over increasingly wide time intervals (Placido et al., 2023) by leveraging data encoded in the temporal sequence of clinical events or "disease pathways" (Placido et al., 2023). The fundamental goal of this study is to apply a Random Forest (RF) model, for use in "computational efficiency, simple interpretation, and high accuracy" (Wang et al., 2023), in a telemedicine web application to create a "scalable workflow for detecting cancer" (Placido et al., 2023).