Systems and methods for adaptative training of machine learning models

Abstract

A system for adapting a machine learning model for a specific population. The system includes a processor and memory devices storing instructions that configure the memory devices to perform operations. The operations may include receive a local dataset comprising local records of patients associated with the healthcare facility, perform a clustering function, and retrieving a template dataset comprising template records organized in clusters with variable centroids. The operations may also include calculating a similarity metric between the local and template records, generating a synthetic dataset by combining template and local records, segregating the synthetic dataset into a training synthetic dataset and a testing synthetic dataset, and generating and/or validating a machine learning predictive model by tuning a template model according to the training synthetic dataset and/or generating a new predictive model.