Data system for evaluating the efficacy of chemoradiotherapy for colorectal cancer and for early warning of adverse reactions

Abstract

The present invention relates to the technical field of medical data processing, and discloses a data system for evaluating the efficacy of chemoradiotherapy for colorectal cancer and for early warning of adverse reactions, comprising a data acquisition module, a data preprocessing module, an efficacy evaluation module, an adverse reaction early warning module, a prediction model module, a visualization and interaction module, and a data storage and management module; by integrating radiomics features (T2-MRI images) and biomarker data (PanCK, CD20, MHC-II), and in combination with a cascaded Transformer network to achieve feature complementarity, patient data are monitored in real time and combined with an early warning model, so that risk warnings can be issued before adverse reactions present obvious clinical symptoms, enabling medical personnel to intervene in advance, thereby reducing the incidence rate of severe adverse reactions and improving the patient treatment experience and the generation of safety early warning information.