References[1] Bevers K., et al. “The Biopsychosocial Model of the Assessment, Prevention,

and Treatment of Chronic Pain”. US Neurology 12.2 (2016): 98-104[2] Bach, K., et al., "Case Representation and Similarity Assessment in the

selfBACK Decision Support System," Cham, 2016, Springer pp. 32-46

AcknowledgmentsThe authors would like to thank the European ResearchCouncil (Back-UP, ID 777090) for providing financial supportto this project.

Milton Hoz de Vila1, Alejandro F. Frangi11Center for Computational Imaging & Simulation Technologies in Biomedicine (CISTIB), University of Leeds, UK

The Back-UP Project: Towards predictive diagnosis and personalised medicine for NLBP underpinned by the VPH technologies and vision

The EU funded Back-Up research project (www.backup-project.eu) will develop, integrate, and validate, patient-specificmultifactorial prognostic models to support effective and efficient management of non-specific Neck and Low Back Pain (NLBP),a complex health problem with major impact on healthcare and worker disability.

NLBP is a dynamic process where biological, psychological,social and economic factors change over time followingcomplex patterns1. Information and statistical models fromthose multidisciplinary dimensions will be integrated into acomprehensive unified system (Fig.1) able to deliver a reliableprognostic model, based on the digital representation ofmultidimensional data, ranging from the molecular level(glycomics) to musculoskeletal structures, integrating as well,the broader psychological, socioeconomic and lifestyle context.

ResultsThe ultimate assessment-intervention modelwill be individually tailored for each patient withchronic pain. Patients, clinicians, insuranceentities and prevention services, will benefitfrom the proposed solution, which approachgoes beyond the traditional scope of medicine,also encompassing, social and occupationalscience.Within this project, MULTI-X will be furtherenhanced and will continue to support scientistsby providing a general-purpose, scalable, andintegrative computational framework (Fig. 4).

MULTI-X, an ecosystem for biomedical researchAll Back-UP heterogeneous datasets, models, and analysispipelines will be integrated into MULTI-X, a cloud-based cross-domain research-oriented platform, designed for collaborative,and reproducible science (Fig. 2).

Introduction

Materials and Methods

With a combination of mechanistic anddata-driven approaches, Back-UP hasthe potential to extend in-silicotechniques to the level of virtual clinicaltrials to produce: (i) personalisedprognostics on the probability ofrecovery, (ii) risk of recurrence, (iii) costanalysis of treatment and return towork timeframe, and (iv) simulations ofaffected musculoskeletal function.Quantifiable results will be reported insubsequent studies.

MULTI-X (www.multi-x.org) will be the underlying core ofthe system. An integration with SelfBACK2 (www.selfback.eu)will incorporate records of the patient’s physical activitiestracked by wearable sensors, and self-assessment datasubmitted using a mobile application. All these resources willbe made available through tight integration with ANTARIHome Care (www.gmv.com), an existing e-health platformthat will be the portal for patients, clinicians, and employersto track and support personalised plans of intervention forrecovery and adaptation of workplace.

2006 2012 2013

Platform Development

2018

Conclusion and Future Work

MULTI-X was designed to foster development, integration, andtesting of scientific tools, facilitating their connection with large-scale federated data repositories and providing analysis tools toget immediate insights about the data processed. It integrateshighly demanded research services, such as container-basedworkflow execution, big data analytics, on-demand machinelearning tools and data visualisation components. Back-UP willfully leverage the capabilities offered by this comprehensive andmodular platform.

CommunityBespoke layout& subdomains

EXPLOREDATA

WORKFLOW

COLLABORATE

ANALYSE

An Open Marketplace for

Precision Medicine

MULTIX

Catalogue

Secure Access

Data

CatalogueDevelopmentEnvironment Tools

CatalogueExecution &MonitoringDesign

Catalogue

Visualisation

Dashboards

ScalableResources

MULTI-X

COLLABORATIVEDesigned for sharing

and co-creationUSER FRIENDLYSimplified development and

deploymentSCALABLE

Fully-managed on-demand resources

CROSS-DOMAINSuitable for multi-

disciplinary use

SECUREControlled and safeguarded

framework

OPENBuilt for

cooperation

INTEROPERABLECapable to integrate existing platforms

DEVELOPMENTenvironment

EXECUTIONenvironment MARKETPLACE

AI SOLUTIONS

INTEGRATIONwith existing systems or

Open Cloud platforms

Fig. 2 Diagram of components and screenshots of the corresponding areas of MULTI-X

This computational and data-intensive analysis platform is theoutcome of more than a decade of collaboration withindifferentVPH-related projects (Fig. 3).

Fig. 3 EU Projects involved in the conception and development of MULTI-X

Fig. 4 MULTI-X addresses the complete processof co-creation of data analytics solutions.

See it in action

Fig. 1 Diagram of the integration of the different components of the Back-UP Platform

M U L T I X

ANALISE

APPLICATIONSMARKETPLACE

COLLABORATE

ACCESSCONTROL

RESULTSVISUALISATION

EXPLORE

WORKFLOWCOMPOSITION

WORKFLOW

DATA

DATASETSMARKETPLACE

MULTI-X

USER INTERFACE &VISUALISATION

Patients & Physicians

GDPRAssessment

module

Data Collection

Data controller

Data QualityAssessment

module

Data curator

CommonHarmonised

Model

Homogenised and verified

data ingestion

Existing Schemas

CEN/ISOarchetypes

HL7C-CDA

HL7FHIR

CDISCSHARE DATA INJECTION &

HARMONISATIONSERVICES

Original Dataset

Content and Terminology

MappingServices

Patientdata

Integration and Harmonisation

SelfMonitoring

MODELING &ANALYTICSSERVICES

Psycho-social

models

Patient Stratification

PatientSelection

Bio-mechanical

models

Researchers & Developers

Other…

OPEN DATASERVICES

Community

Socio-economic

models

PrognosisRisksCosts