Early temporal prediction of Type 2 Diabetes Risk Condition from a General Practitioner Electronic Health Record: A Multiple Instance Learning Boosting Approach

published in Artificial Intelligence in Medicine by M. Bernardini, M. Morettini, L. Romeo, E. Frontoni and L. Burattini:

@article{bernardini2020early, title={Early temporal prediction of Type 2 Diabetes Risk Condition from a General Practitioner Electronic Health Record: A Multiple Instance Boosting Approach}, author={Bernardini, Michele and Morettini, Micaela and Romeo, Luca and Frontoni, Emanuele and Burattini, Laura}, journal={Artificial Intelligence in Medicine}, pages={101847}, year={2020}, publisher={Elsevier} }

The aim of this study was to propose a multiple instance learning boosting algorithm, named MIL-Boost. The MIL-Boost was applied to past electronic health record patients' information stored by a unique General Prctitioner in order to create a predictive model capable of early predicting insulin resistance worsening (low vs high T2D risk) in terms of TyG index.

We tested the predictive performance of the MIL-Boost approach on the Italian Federation of General Practitioners dataset, named FIMMG_pred dataset, publicly available at the following link: http://vrai.dii.univpm.it/content/fimmgpred-dataset

Run MILBoost.m file to replicate the MIL-Boost experimental procedure.

The authors have been inspired in part by the Multiple Instance Learning algorithm developed by David M. J. Tax (https://github.com/DMJTax/mil).