Editorial: The use of data mining in radiological-pathological images for personal medicine

Abstract

Editorial on the Research TopicThe use of data mining in radiological-pathological images for personal medicineThe use of data mining in radiological-pathological images for personalized medicine is a promising and rapidly developing field. This innovative approach involves the integration of large amounts of data from medical images, pathology reports, and clinical records to improve patient care and treatment outcomes.Currently, there are mainly two methods for data mining on medical images: image findings (subjectively summarized factors) (Renzulli et al., 2016;Yue et al., 2022) and high-order features (objectively calculated features, such as radiomics and pathomics) (Lambin et al., 2017;Chen et al., 2022). Image findings can be easily obtained and explained by pathophysiological mechanisms (Segal et al., 2007). However, they are determined by readers and stability need to be improved. Fortunately, deep learning techniques can behave like an experienced reader and compute more robust evaluation results (Zhao et al., 2020). High-order features depend on image analysis techniques which attracted great attention in the last few years. Most research focused on tumor characteristics and prognosis (