Cloud-Enabled Predictive Modeling of Cancer Progression in Digital Twins: A LightGBM Classification Approach

Cancer patients have witnessed an advancement in diagnostic capabilities based on the accumulation of progressive data. It enables the categorization and analysis of diagnosis based on countless internal factors. The patient data can be securely housed on cloud-based platforms, affording authorized medical practitioners and experts a seamless means of access and reference. The utilization of cloud platforms not only facilitates the storage of day-to-day patient data but also facilitates the addition of post-diagnostic information. The cloud environment exhibits exceptional qualities, such as high scalability, adaptable backup storage solutions, efficiency in resource optimization, and effortless service acquisition. In our efforts to leverage this valuable repository of patient data and offer users, both current and future, with comprehensive diagnostic insight, the concept of Digital Twins (DTs) is employed which is an innovative approach that allows individuals sharing similar cancer diagnosis or prognoses to explore potential remedies by drawing upon the collective knowledge in cloud. The replication and analysis of this data are orchestrated through a sophisticated machine learning algorithm, primarily Light Gradient-Boosting Machine (LightGBM), which incorporates elements of Extreme Gradient Boosting (XGBoost). Together with Digital Twins (DTs), this amalgamation substantially enhances the accuracy of diagnostic predictions for the given dataset. It empowers both patients and healthcare professionals with a cutting-edge tool for more precise and personalized cancer management.