Realistic simulation of virtual multi-scale, multi-modal patient trajectories using Bayesian networks and sparse auto-encoders

• Published in: Scientific reports Vol. 10; no. 1; p. 10971
• Main Authors: Sood, Meemansa, Sahay, Akrishta, Karki, Reagon, Emon, Mohammad Asif, Vrooman, Henri, Hofmann-Apitius, Martin, Fröhlich, Holger
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 03.07.2020; Nature Publishing Group
• Subjects: 692/308/153; 692/308/575; Alzheimer Disease - diagnostic imaging; Alzheimer Disease - genetics; Alzheimer's disease; Bayes Theorem; Bayesian analysis; Brain - diagnostic imaging; Brain research; Cognitive ability; Cohort Studies; Computer Simulation; Databases, Factual - statistics & numerical data; Deep Learning; Disease Progression; Drug development; Humanities and Social Sciences; Humans; Longitudinal Studies; Mathematical models; Medical imaging; Models, Statistical; multidisciplinary; Parkinson Disease - diagnosis; Parkinson's disease; Patients; Polymorphism, Single Nucleotide; Science; Science (multidisciplinary); Simulation; Statistical models; Translation; Translational Research, Biomedical - methods; Translational Research, Biomedical - statistics & numerical data; User-Computer Interface
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-020-67398-4

Translational research of many disease areas requires a longitudinal understanding of disease development and progression across all biologically relevant scales. Several corresponding studies are now available. However, to compile a comprehensive picture of a specific disease, multiple studies need to be analyzed and compared. A large number of clinical studies is nowadays conducted in the context of drug development in pharmaceutical research. However, legal and ethical constraints typically do not allow for sharing sensitive patient data. In consequence there exist data “silos”, which slow down the overall scientific progress in translational research. In this paper, we suggest the idea of a virtual cohort (VC) to address this limitation. Our key idea is to describe a longitudinal patient cohort with the help of a generative statistical model, namely a modular Bayesian Network, in which individual modules are represented as sparse autoencoder networks. We show that with the help of such a model we can simulate subjects that are highly similar to real ones. Our approach allows for incorporating arbitrary multi-scale, multi-modal data without making specific distribution assumptions. Moreover, we demonstrate the possibility to simulate interventions (e.g. via a treatment) in the VC. Overall, our proposed approach opens the possibility to build sufficiently realistic VCs for multiple disease areas in the future.