Time-to-Event Interpretable Machine Learning for Multiple Sclerosis Worsening Prediction: Results from iDPP@CLEF 2023

iDPP CLEF 2023 - Conference and Labs of the Evaluation Forum - -2023

Authors

Lombardi Angela, De Bonis Maria Luigia Natalia, Fasano Giuseppe, Sportelli Alessia, Colafiglio Tommaso, Lofù Domenico, Sorino Paolo, Narducci Fedelucio, Di Sciascio Eugenio, Di Noia Tommaso

Abstract

Background. A variety of symptoms have been reported, but the prevalence of specific symptoms in relapsing-remitting multiple sclerosis (RRMS), how they are related to one another, and their impact on patient reported outcomes is not well understood.Objective. To describe how symptoms of RRMS cooccur and their impact on patient-reported outcomes.Methods. Individuals who reported a physician diagnosis of RRMS in a large general health survey in the United States indicated the symptoms they experience because of RRMS and completed validated scales, including the work productivity and activity impairment questionnaire and either the SF-12v2 or SF-36v2. Symptom clusters were identified through hierarchical cluster analysis, and the relationship between clusters and outcomes was assessed through regression.Results. Fatigue, difficulty walking, and numbness were the most commonly reported symptoms. Seven symptom clusters were identified, and several were significantly related to patient reported outcomes. Pain, muscle spasms, and stiffness formed a cluster strongly related to physical quality of life; depression was strongly related to mental quality of life and cognitive difficulty was associated with work impairment.Conclusions. Symptoms in RRMS show a strong relationship with quality of life and should be taken into consideration in treatment decisions and evaluation of treatment success.

DOI

https://doi.org/

BibTex references

@Article{LDFSCLSNDD23,
  author       = "Lombardi, Angela and De Bonis, Maria Luigia Natalia and Fasano, Giuseppe and Sportelli, Alessia and Colafiglio, Tommaso and Lof\`u, Domenico and Sorino, Paolo and Narducci, Fedelucio and Di Sciascio, Eugenio and Di Noia, Tommaso",
  title        = "Time-to-Event Interpretable Machine Learning for  Multiple Sclerosis Worsening Prediction: Results from iDPP@CLEF 2023",
  journal      = "iDPP  CLEF 2023 - Conference and Labs of the Evaluation Forum ",
  year         = "2023",
  url          = "http://sisinflab.poliba.it/Publications/2023/LDFSCLSNDD23"

}