In this work, we present a framework for the interpretable analysis of machine learning algorithms to predict the Multiple Sclerosis worsening using the datasets provided by the iDPP@CLEF 2023 Challenge. The proposed framework is modular and allows to investigate the link between the provided static and dynamic features and the outcome to be predicted. Our findings show that better performance could be achieved by using Random Survival Forests together with temporal information about the clinical scores and a proposed feature related to the normalized frequency of patients’ relapses.
Time-to-Event Interpretable Machine Learning for Multiple Sclerosis Worsening Prediction: Results from iDPP@CLEF 2023 / Lombardi, A.; De Bonis, M. L. N.; Fasano, G.; Sportelli, A.; Colafiglio, T.; Lofù, D.; Sorino, P.; Narducci, F.; Di Sciascio, E.; Di Noia, T.. - 3497:(2023), pp. 1272-1285. (Intervento presentato al convegno 24th Working Notes of the Conference and Labs of the Evaluation Forum, CLEF-WN 2023 tenutosi a grc nel 2023).
Time-to-Event Interpretable Machine Learning for Multiple Sclerosis Worsening Prediction: Results from iDPP@CLEF 2023
Lombardi A.;De Bonis M. L. N.;Sportelli A.;Colafiglio T.;Lofù D.;Sorino P.;Narducci F.;Di Sciascio E.;Di Noia T.
2023-01-01
Abstract
In this work, we present a framework for the interpretable analysis of machine learning algorithms to predict the Multiple Sclerosis worsening using the datasets provided by the iDPP@CLEF 2023 Challenge. The proposed framework is modular and allows to investigate the link between the provided static and dynamic features and the outcome to be predicted. Our findings show that better performance could be achieved by using Random Survival Forests together with temporal information about the clinical scores and a proposed feature related to the normalized frequency of patients’ relapses.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
