GP Progression Model¶

The Gaussian Process Progression Model (GPPM) is a model of disease progression estimating long-term biomarkers’ trajectories across the evolution of a disease, from the analysis of short-term individual measurements. GPPM software has been first presented in the work Lorenzi, NeuroImage 2017, and subsequently extended in the GPPM-DS presented in Garbarino and Lorenzi, IPMI 2019 and Garbarino and Lorenzi, NeuroImage 2021.

GPPM and GPPM-DS enable the following analyses:

  • [GPPM] reoconstruct the profile of biomarkers evolution over time,

  • [GPPM] quantify the subject-specific disease severity associated with the measurements each individual (missing observations are allowed),

  • [GPPM] estimate the ordering of the biomarkers from normal to pathological stages,

  • [GPPM-DS] specify prior hypothesis about the causal interaction between biomarkers,

  • [GPPM-DS] data-driven estimation of the interaction parameters,

  • [GPPM-DS] data-driven comparison between different hypothesis to identify the most plausible interaction dynamics between biomarkers,

  • [GPPM-DS] model personalisation to simulate and predict subject-specific biomarker trajectories,

Getting started

The software comes with a simple installation and an easy interface.

An example of the basic usage of GPPM on synthetic and real data is available here:

[Basic GPPM tutorial]

[Jupyter notebook]

[Colab notebook]

An example of GPPM-DS on synthetic and real data is available here:

[Basic GPPM-DS tutorial]

[Jupyter notebook]

[Colab notebook]


The source code is available on GitLab. The software is freely distributed for academic purposes. All the commercial rights are owned by Inria.