Risk Prediction
as a Service:

a DSS architecture promoting
interoperability and collaboration

Stefano Mariani

Università di Modena e Reggio Emilia


  1. Scenario
  2. Proposition
  3. Enabling technologies
  4. Application


  • Growing ICT impact on clinical practice and research
    • ML >> predictive modelling >> decision support
    • i.e. risk prediction
  • Barriers to overcome
    • standards
    • technological development

Use cases

  1. Data science teams want to join efforts for better predictive models
    • geographical distribution >> remote collaboration only
    • different organisations >> data sharing cumbersome
    • different technical skills >> sharing code barely useful

  1. Clinical staff wants to experiment predictive modeling
    • lacks IT resources >> up-front investment
    • lacks IT skills >> steep learning curve

  1. Healthcare organisations want to assess transferability of models
    • lack reciprocal trust >> no data disclosure
    • different technical skills >> incompatible models


Ubiquitous access & seamless deployment

of prediction models.

Interoperability & collaboration

between data science teams and, possibly, clinicians.


DSS architecture

Web API: RESTful endpoints

Translation: R/Python models to PMML/PFA

Prediction: Serve predictions

Learning: Build prediction models

Storage: Store models


  • Plugin mode:
    • upload model
    • list models
    • inspect model (metadata)
    • download model
    • delete model
    • apply model
  • Learning mode:
    • train model
    • test model
    • save model (freeze)
    • discard model

  • Common:
    • set global (usable anywhere vs. provider site)
    • set online (learning on new data vs. "frozen")

Non-functional Properties

Trust: local validation + transparency (inspection)

Language / framework agnostic: automatic translation + PMML/PFA (R, caret + Python, scikit-learn)

Interoperability: PMML/PFA, HTTP API, JSON/XML data

Accessibility: Web Service, REST architecture

Enabling technologies

  • "X-as-a-Service" paradigm
    • especially: "ML-as-a-Service" (MLaaS)
  • Standards promoting seamless exchange and deployment of ML models
    • don't brake the data pipeline
    • language / framework agnostic >> "build once, run everywhere"

Service Orientation

  • "as-a-Service" paradigm to deliver software products
    • roots in Service Oriented Architecture (~90s)
    • well-known implementation: Web Services (~2000s)
    • modern evolution: RESTful micro-services (~2010s)
  • MLaaS growing trend
    • i.e. IBM Watson ML, Amazon SageMaker, MS Azure ML Studio, Google Cloud ML Engine
    • off-the-shelf solutions : wide portfolio of models, customisation options, user friendly
    • lack focus on serving models (i.e. IBM MAX)


  • Predictive Model Markup Language
    • ML models representation format based on XML

  • Portable Format for Analytics
    • ML pipelines representation format based on JSON

independent development / deployment of predictive models

flexibility of production environment

co-existence of different programming languages and frameworks


  • Architecture implemented in CONNECARE European project
    • Predictions about hospitalisations, emergency re-admissions, mortality
    • Models created by partners with locally available data (6000+ patients with 5+ years follow-ups)
  • DSS as precious asset to let clinical partners
    • use and validate each other models
    • cooperatively improve models without data sharing
    • assess transferability

Risk Prediction

CONNECARE readmission risk prediction use case

Home Hospitalisation (HH) / Early Discharge (ED)

  • During HH / ED (t0) >> stratify patients to optimise care (Risk Models 1-3)
    • risk of early readmission after hospital discharge
    • risk of mortality
  • after HH / ED (t1) >> stratify patients for transitional care (Risk Models 4-6)
    • risk of early readmission after hospital discharge
    • risk of mortality


Reference architecture for a DSS, promoting

  • interoperability amongst technological environments
  • collaboration between teams of data scientists, and towards clinical staff

State of art tech + ML standards = support to practical use cases

  • service-orientation, REST architecture
  • PMML and PFA data formats

Running implementation within CONNECARE

  • clinical case studies ending after summer >> assessment of DSS efficacy
  • CONNECARE ending Dec 2019 >> DSS released as open-source


for your attention


Stefano Mariani

Università di Modena e Reggio Emilia