Abstract: To automatically model oncology treatment severity levels for a particular patient, a plurality of independently generated and potentially inconsistent observation data records are received for a patient. A condition-specific plurality of observation data records each having a common cancer type identifier are selected, and those condition-specific observation records are filtered to potentially eliminate one or more observation data records failing to satisfy one or more preliminary filter criteria. A specific pre-processing process for the remaining observation data records is executed to generate one or more output data records having at least one shared identifier, and the output data records are provided as model input to a machine-learning based severity model to generate a severity score for the patient.

Abstract: To enable automated processing of certain observation data by machine-learning based models, as well as to ensure those machine-learning based models are trained utilizing reliable and consistent data, independently generated observation data records each comprising biomarker mutation indicators, are utilized to generate a model input data set by applying biomarker mutation indicator-based filters, including intra-date filters and inter-date filters to identify and rectify observation data records that do not satisfy applicable biomarker mutation indicators. By identifying and rectifying the observation data records that do not satisfy applicable biomarker mutation indicators, a clean, model input data set is generated that is then utilized to generate a severity score for the patient.