Table 3.

Research overview: prognosis with statistical models.

Key	Dataset(s) origin	Dataset(s) type	Dataset(s) length	Scope	Biomarker(s) type	Pre-processing (if any)	Validation (if any)	Model(s) tested
Ince et al., 2003	Newcastle upon rajiTyne MND clinic	Real-life	81	Progression	Imaging	None described	Not required	Univariate analysis
Forbes, 2004	Scottish ALS-MND Register	Population	1226	Outcome	Clinical	None described	Not required	Cox time rajidependent regression modeling
Visser et al., 2007	Dutch university hospitals	Real-life	37	Outcome	Clinical, genetic, biological	None described	Not required	Univariate analysis
Coon et al., 2011	Mayo Clinic	Real-life	56	Outcome	Clinical, imaging	None described	Not required	KM analysis
Atassi et al., 2014	PRO-ACT	Clinical trial	8635	Outcome, progression	Clinical, biological	Data cleaning	Not required	Multivariate analysis
Elamin et al., 2015	Irish and Italian (Piemonte) ALS registry	Population	326	Outcome	Clinical, genetic	FS	HOV	Proportional hazards Cox
Marin et al., 2015	FRALim register	Population	322	Outcome	Clinical	None described	Not required	Cox regression (KM)
Rong et al., 2015	-	Clinical trial	66	Progression	Clinical	FS	Not required	Linear Mixed Effect, KM analysis
Tortelli et al., 2015	University of Bari MND Center	Clinical trial	145	Outcome	Clinical	None described	Not required	Bivariate model for correlation
Wolf et al., 2015	Rhineland-Palatinate Register	Population	193	Outcome	Clinical	FS	Not required	Cox proportional hazards
Knibb et al., 2016	South-East England Register	Population	575	Outcome, progression	Clinical	MVR	CV	Cox proportional hazards, ACT
Reniers et al., 2017	University Hospitals Leuven	Real-life	396	Outcome	Clinical	None described	Not required	Univariate and multivariate Cox regression

HOV, Hold Out Validation; CV, Cross Validation; ACT, Accelerated Failure Time; KM, Kaplan Meier; MVR, Missing Value Removal; FS, Feature Selection.