Taotlus

Project title: Health Sense Project Promoter: Estonian Ministry of Social Affairs Project Partner(s): Health and Welfare Information

Systems Center, Tallinn University of Technology, University of Tartu

Donor project partner(s): Helse West IKT; Western Norway University of Applied Sciences

Total maximum eligible project cost: 2 000 000€

Project grant rate: 100%

Project grant amount: 2 000 000€

Estimated duration: 39 months

a) Project summary

The goal of the predefined project Health Sense is to develop a secure data storage, integration,

access and analysis toolkit in order to provide large, complex and detailed sets of health and

lifecycle data for the public sector, private sector and R&D institutions. The ultimate aim of the

project is to raise the availability of health and lifecycle data in order to enhance the creation of

new products, services and solutions by the private sector, novel interventions by the public

sector and relevant scientific knowledge by the R&D communities in order to raise life

expectancy and prolong healthy life years of people. It is accounted for, that while the user

community is likely to broaden even further in the future, then this data is delicate, which means

privacy and security of developed/procured components is key. However in addition to

technical security, compliance to strict data protection regulations is crucial as well. Thus, a

key activity stream will aim to develop data collection procedures, tools and analyse regulations

that meet limitations and responsibilities described in the General Data Protection Regulation1

(GDPR) to ensure usability of developed ICT components and thus ensure project success.

Health Sense is a virtual environment used to collect health, health cycle and lifestyle data for

research and service development purposes. In addition, it also provides a toolbox that enables

the creation of data analysis-based solutions for prolonging a healthy human life. As the data

handled in the environment is sensitive, the tools for safe collection, storage, access and

processing of data are an essential part of the provided toolbox.

b) Project background and justification

Estonian governmental information system today has sufficiently good e-ability to collect and

handle personal data in different areas including healthcare, social care and labor area that are

managed by Ministry of Social Affairs. Mostly these abilities today are used on an operational

level to provide services for individuals, and for reporting and simple analyses. The better use

of data, to provide the steering process with predictive and calculated supporting data, is

1 Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of

natural persons with regard to the processing of personal data and on the free movement of such data, and

repealing Directive 95/46/EC. OJ L 119, 4.5.2016, p. 1–88.

planned but not yet implemented. Today a couple of domain-based separated data warehouses

are used, and data relations between different areas are built, but only on a data mart level. This

leads to limited possibilities in data analysis and AI development.

Today many different countries have the same problem – services are being provided and data

is being collected, but separately and by areas with very simple loop back to steering process.

Even when the data collected by different areas is gathered to the same environment, the data

from different sources is connected on a person level and no additional relations are calculated

nor defined. In many circumstances, legal constraints prevent interconnecting person level data.

Estonia is small and dynamic enough to build data integration, analysis models, and solutions

to experiment with data warehouse environment architecture and development tools. We have

all the necessary electronic data sources for supporting this process. Furthermore, we have years

of experience in developing forward-thinking digital solutions and in simultaneously

building/updating our regulatory framework for these solutions to be implemented. Estonia,

being the near perfect testbed for spearheading new e-governance initiatives, can provide here

great value by developing models and solutions, which can later be expanded further to larger

data volumes in larger countries, Norway being one of the main considerations.

It is paramount to find committed partners, that are interested in projects of this kind as well as

who comply with ethics and the principles of data security. There are several interested groups

in the public sector, R&D institutions and private sector. The Department of Data Analysis has

been set up under the Ministry of Social Affairs, however the need for health and lifecycle data

continuously exists in other public sector institutions in the health and social domain (e.g.

Health Insurance Fund, Health Development Institute, Estonian Health Board, Social Insurance

Board, Unemployment Insurance Fund) and other public sector domains as well (e.g. Ministry

of Defense, Defense Resources Agency, Ministry of Interior, Ministry of Education and

Research etc). R&D institutions (e.g. University of Tartu and Tallinn University of Technology

and Technology Competence Centers) are interested in using humans’ lifecycle data in their

research, development and teaching processes, in cooperation with the governmental

authorities. Various lifecycle and health data interested communities (e.g. Estonian HealthTech

Cluster, Open Knowledge Estonia) are set up so that they connect individuals, research

organizations, companies and startups. In addition, in recent years international interest by

global (multi-national) companies (mainly pharmaceuticals and health technology) has grown

exponentially.

All these groups are interested in having access to person’s lifecycle data. Some of them

produce human lifecycle data, and through that expand the variety of data collected. Due to the

high level of IT coverage in Estonia, many international companies have become interested in

using health and lifecycle data in their R&D efforts. This can provide value as additional

resources to the project.

Implementing the modern methods of person’s lifecycle-based data collecting, analysis and AI

tools will lead to a longer and healthier life and reducing costs per unit in health- and social

care. Thus, supporting the goal of the welfare technologies concept – staying independent in

activities of daily living while receiving tailored support from various care services when

needed.

Health Sense is a central (root) project for the redevelopment of data utilization in health, work

and social care area, where the polices and services are intertwined and thus should be combined

to provide maximum effectiveness and cost-efficiency. The latter is currently hindered by the

scatteredness of data in these areas which leads to deficiencies in accessing and utilizing data

for revising and building services. Thus, the toolkit developed through Health Sense would

enable to modify, create and provide invisible and/or preventive services to citizens based on

evidence gather through complex data analysis/findings. The project will be implemented as a

predefined project because it requires the coordination between several public and private

actors, many of which have very specific knowledge in the project area. Without the successful

implementation of this project today, the push for combined evidence and data-led

improvement of human social and health wellbeing, will be delayed for years.

c) Objective, expected outcome(s) and list of outputs

Project activity streams:

1.1. Project management

1.2. Data collecting environment

1.3. Data storage with pseudonymization and obfuscation tools

1.4. Warehouse data structures and ETL process

1.5. Data Analysis and visualization environment for third parties (e.g. private sector, R&D

institutions)

1.6. Building data flows in cooperation with third parties (e.g. private sector, R&D institutions)

1.7. Analysis of legal regulations to foster the results of the project

For more detailed activities see timetable/milestone plan (e). For budget overview see sheet

accompanied with this document.

Expected outputs:

 data collection environment from different sources (governmental and non-governmental)

 data pseudonymization, data serialization/deserialization, obfuscation and long text

extraction tools

 partly encrypted data storage with encryption/decryption tools and methodology

 data flows from runtime information system to data warehouse (built on needs of third

parties). Illustrated with more detail in Appendix 2 and 3.

 user rights and consent management system

 data analysis environment for third parties

 Analysis of legal regulations to ensure project outcomes

 data warehouse environment for Ministry of Social Affairs and other governmental

institutions (for reporting and analysis purposes)

Expected outcomes:

 to create a data management platform for developing innovative solutions for healthcare

and welfare service providers to develop and experiment new innovative products and

services;

 to encourage the emergence of new health, welfare and lifestyle businesses and data driven

business models;

 to create a launch pad for new healthcare and welfare services for their emergence to the

market in cooperation with SMEs and start-ups;

 to promote scientific thinking and provide new inputs for research, development and

innovation in SMEs and start-ups in the delivery of health and welfare services and educate

highly skilled workforce in cooperation with universities.

Expected

project

result

Indicator

Unit of

measure

ment

Source of

verification

Frequency

of reporting

Baseline

value

Baseline

year

Target

value

Health

Sense data

management

platform

developed

and

operating

Data analysis

procedures/too

ls in place and

documented

Binary

Documentation

available on

HWISC website,

software hosted

at HWISC or

state cloud

infrastructure

Semi-

annually No 2018 Yes

Data

collection and

publishing

procedures/too

ls in place and

documented

Binary

Documentation

available on

HWISC website,

software hosted

at HWISC or

state cloud

infrastructure

Semi-

annually No 2018 Yes

Data

collection, and

publishing

environment

architecture

and data

analysis

environment

architecture

developed and

published

Binary

Documentation

available on

HWISC website,

software hosted

at HWISC or

state cloud

infrastructure

Semi-

annually No 2018 Yes

Number of

data

mining/Artific

ial Intelligence

pilot projects

successfully

finished

Number

Documentation

outlining

projects and

results available

on HWISC

website

Semi-

annually 0 N/A 3

Number of

professional

staff trained

Number Attendance

sheets

Semi-

annually 0 N/A 10

Project sustainability

All results created during the project will be maintained by Health and Welfare Information

Systems Centre and Ministry of Social Affairs. All software created during the project will be

hosted and maintained by Health and Welfare Information Systems Centre. The necessary

resources to maintain and host results will be applied from the annual state budget of Estonia.

Intellectual property

Unless stipulated otherwise in the agreements between project promoter and project partners,

intellectual property rights of the results of the project shall be owned by the party generating

it. When results are generated jointly, it shall be jointly owned, unless the parties in question

agree on a different solution. The method of defining joint ownership shall be based on their

contributing efforts towards the result in shares proportionate to their scope of work/budget for

that result.

After completion of the project, project promoter shall have the non-exclusive, royalty-free

right to use, modify, copy, transfer or make derivative results of the project. The rights of the

project promoter shall not be limited with regard to certain rights or to the purpose, term,

territory, extent, manner or means of using the results of the project.

d) Information on the project promoter and project partner(s)

Ministry of Social Affairs is a governmental institution that operates in the field of

social security. For ministry have been set five strategic objectives - ensure people's

economic prosperity and their good work, ensure people's social coping and

development, support the well-being of children and families, promote people's

mutual care, equal opportunities and gender equality, and ensure people's long and

high-quality life. Ministry of Social Affairs emerges objectives through compiling

development plans for the fields being under control of ministry, as well as

organization-based development plans.

Ministry of Social Affairs as project promoter, will ensure seamless project

coordination.

Health and Welfare Information Systems Centre (HWISC) is info technology

competence centre of Ministry of Social Affairs. HWISC is involved in the

development, and management of ICT services in the health, social and labour fields

within the area of government of the Ministry of Social Affairs. HWISC is a

representative of the Ministry of Social Affairs in the „Health Sense“ project as the

project manager.

HWISC will take upon the leading role in most of the project activities including

procuring the necessary licensed software and hosting as well as designing the

technical architecture of components and maintaining the solutions developed

during Health Sense project. HWISC will assist in project coordination and in

organizing meetings as well as in analysis of legal regulations to ensure the success

of the project.

Tallinn University of Technology (TalTech), the second largest university in

Estonia, is international university teaching the knowledge and skills necessary for

today’s international and competitive marketplace. TalTech is the top centre of

science in Estonia. With strong emphasis on internationalization, TalTech has a

strong multicultural student body, many international professors (11% foreign staff

members), great cooperation with the world's top universities (e.g. Stanford,

Technische Universität München, Del University of Technology and KTH Royal

Institute of Technology in Europe) and with international companies (e.g.,

Microsoft, ABB, IBM. etc.). There are over 30 fully accredited international degree

programmes (4 Bachelor programmes, 18 Master programmes and 10 PhD

programmes) that are available fully in English.

TalTech will develop the appropriate model for digital health care data that could

be used for data collection and storage in new generation Estonian Health

Information System with the feasibility to implement the same model in other large

scale Electronic Health Records (EHR). TalTech will develop the data model for

digital medical data capture and storage to achieve semantic interoperability and

develop universal continuity of care process standard.

University of Tartu (UT) is Estonia’s leading centre of research and training. It

preserves the culture of the Estonian people and spearheads the country's reputation

in research and provision of higher education. UT belongs to the top 1.2% of world's

best universities.

As Estonia's national university, UT stresses the importance of international co-

operation and partnerships with reputable research universities all over the world.

The robust research potential of the university is evidenced by the fact that the

University of Tartu has been invited to join the Coimbra Group, a prestigious club

of renowned research universities.

UT includes four faculties. To support and develop the professional competence of

its students and academic staff, the university has entered into bilateral co-operation

agreements with 79 partner institutions in 31 countries.

University of Tartu will undertake the development of obfuscation toolkit, data

serialization/deserialization tool analysis and POC.

Helse West IKT and Western Norway University of Applied Sciences are

cooperating in building infrastructure for Intromat (https://intromat.no/), a project

on introducing e-health in mental health. As Intromat requires designing of

architecture, API-s, interfaces, pseudonimization/synonimization tools similar to the

aim of Health Sense proposal, then it provides an ample opportunity to build

synergies for both projects and future collaborations. Helse Vest IKT supplies all

IT services to specialized health services in the counties of Rogaland, Hordaland

and Sogn og Fjordane. Helse Vest customers include the public hospitals in Helse

Vest, and several private health service providers in the region. Both Helse Vest IKT

and Western Norway University of Applied Sciences have extensive experiences in

designing and building software architecture.

Bilateral partners will provide consultations and best practice exchange on the

following topics: warehousing architecture (e.g. component architecture, processes,

relations with other warehouses, ETL principles and relations with data providers,

standards), licensed components (database environment, ETL tools, analytical tools,

data virtualization environments) and developed tools (synonymization tools,

obfuscation tools, tools for long text extraction, data encryption/decryption). The

bilateral partnership is crucial to ensure, that solutions developed in Estonia are

state-of-the-art, internationally relevant and to gauge the potential for sharing

components with Norwegian counterparts.

e) Timetable / Milestone plan *)

Extensive timetable is also visualized in a separate excel sheet

Activity (see attached

timetable)

Activity

stream

Leader Timeframe

(week from

project

starting date)

Project kick-off: 1st February 2021 (or as soon as possible)

Staffing the project

team

1.1 Ministry of Social Affairs, Health and

Welfare Information Systems Centre

1-40

A communication

campaign and

negotiations to select

third party projects

1.1 Ministry of Social Affairs, Health and

Welfare Information Systems Centre

1-170

Building overall

system architecture

1.1 – 1.7 Health and Welfare Information

Systems Centre (incl. project

consortium*)

1-23

Analysis of legal

regulations to ensure

project success

1.7 Ministry of Social Affairs, Health and

Welfare Information Systems Centre

40-102

Selecting and

describing projects for

develompent with third

1.1, 1.5-1.6 Health and Welfare Information

Systems Centre

6-170

parties (private sector,

R&D institutions)

Building hardware

architecture

1.1 – 1.7 Health and Welfare Information

Systems Centre (incl. project

consortium*)

4-9

Describing the criteria

and restrictions for the

Initial software

components to be

procured (ETL, DW

database, PostgreSQL.

Encryption/decryption,

obfuscation,

pseudonymisation,

user permissions,

virtualization etc.

1.1 – 1.7 Health and Welfare Information

Systems Centre (incl. project

consortium*)

1-170

Installation of initial

environment to

infrastructure

1.1 - 1.6 Health and Welfare Information

Systems Centre

8-11

Component,

installation, and testing

(Data collecting

environment – ETL,

BIW tools)

1.1, 1.4,

1.5

Health and Welfare Information

Systems Centre

9-14

Building virtual

working environment

for third parties (closed

environment for ETL,

data marts, and analyse

tools)

1.1, 1.5 -

1.6

Health and Welfare Information

Systems Centre

12-170

Tenders of licenced

tools

1.1 – 1.5 Health and Welfare Information

Systems Centre

9-16

Completing initial

environment (tender

result installations) –

central DW and data

lake environment

1.1 – 1.5 Health and Welfare Information

Systems Centre

17-21

Testing of

environment

1.1 – 1.5 Health and Welfare Information

Systems Centre

22-25

Description of initial

environment, user

manuals

1.1 – 1.5 Health and Welfare Information

Systems Centre

9-25

Development/detailing

of architecture

1.4 Project consortium* 5-150

Building BIW for IS-s 1.1 – 1.5 Health and Welfare Information

Systems Centre

26-90

Development of

logical data structures

for DW

1.4 Tallinn University of Technology 2-64

Building data flows

from BIW to central

DW (ETL)

1.1 - 1.5 Health and Welfare Information

Systems Centre

17-90

Development of user

rights environment

1.1 – 1.5 Health and Welfare Information

Systems Centre

22-34

API, REST etc.

interfaces to access

data

1.5 Health and Welfare Information

Systems Centre

40-170

Third party projects

testing the built

toolkits and

environment

1.1 - 1.6 Health and Welfare Information

Systems Centre

26-170

Development of tools

for collection of

aggregated data

1.5 - 1.6 Health and Welfare Information

Systems Centre

52-170

Environment for

publishing data sets

and statistical results

(Open Data)

1.5 Health and Welfare Information

Systems Centre

58-70

Development of

pseudonymisation

toolset

1.3 Health and Welfare Information

Systems Centre

9-73

Development of

obfuscation toolset,

data

serialization/deserializ

ation tool analysis and

POC

1.3 University of Tartu 9-170

Development of

serialization/deserializ

ation tool

1.3 Health and Welfare Information

Systems Centre

42-170

Meetings with partners

(in Estonia)

1.1 Health and Welfare Information

Systems Centre, Helse-West IKT,

Western Norway University of Applied

Sciences

43, 66, 96

Meetings with partners

(in Norway)

1.1 Helse-West IKT, Western Norway

University of Applied Sciences, Health

and Welfare Information Systems

Centre

56, 76, 86

Final descriptions of

environment, user

manuals

1.1 Health and Welfare Information

Systems Centre

90-170

Promotion of project

results

1.1 Health and Welfare Information

Systems Centre 95-170

End of project: 30th April 2024

*Project consortium includes project promoter and all project partners (incl. donor partners).

f) Budget

Included in separate Excel sheet.

g) Overview of risks:

1. Legal frame obstacles to be followed in order to merge the different datasets.

a. Probability: Likely

b. Impact: Moderate

c. Response: Development of legal framework (incl. revising current regulations)

will go in strong cooperation with the project team and project objectives.

Revising regulations is part of not only major, but also minor e-government

initiatives in Estonia. This is done regularly, routinely and simultaneously for

all new governmental ICT-related initiatives as well as for minor modifications

in existing digital infrastructure.

2. Breech of principles and laws regarding privacy, data safety and ethics.

a. Probability: Unlikely

b. Impact: Major

c. Response: Apply strict data security measures and methods throughout data

storage and transfer infrastructure. Provide third-party access to only

pseudonymized data only within the secure environment with access to

analytical tools. Log all actions, which are done with the data in the secure

environment. Allow only aggregated/analysed data to be exported out of the

secure environment after logs have been checked for any suspicions. Additional

principles regarding data privacy and trust to minimize this risk are detailed in

Appendix 1.

3. Procurements for hardware and/or software fail or will be delayed that leads to difficulty

to support requested activities.

a. Probability: Unlikely

b. Impact: Major

c. Response: Thorough preparation of the procurement documents

4. Lack of interest from SMEs and/or academia that leads to underuse of the Health sense

capability.

a. Probability: Unlikely

b. Impact: Moderate

c. Response: Engage additional entities to promote the capability. Provide

additional operational support from governmental agencies (e.g. datasets that

can be used to solve specific problems).

5. Reputation damage to the project owner and project partners

a. Probability: Unlikely

b. Impact: Minor

c. Response: Carefully plan, prepare and staff the project before the official start

to mitigate planning fallacies. Carefully select partners based on their showcased

skills, interest, past commitments and adherence to data privacy and ethical

principles.

6. Postponement of welfare technology open call projects

a. Probability: Likely

b. Impact: Moderate

c. Response: Include also projects/partners with reliable funding outside of welfare

technology open call in addition to the open call projects.

APPENDIX 1 – Privacy and trust requirements and related responses

1. Requirements

a. The results of the project require the application of high-level data protection and sensitivity

rules to prevent data misuse and privacy breaches.

b. All data processing activities must be verifiable at a later stage.

c. Data processing must take place on the basis of legislation.

d. Only deal with ethical and technically capable partners who can provide a secure data storage

environment and trusted team.

2. Responses

a. Data storages are divided to two securely separated areas: data owner area and researchers

area; personalized data can be stored only at the data owner area; at the researchers area data

can be stored in pseudonymized and aggregated manner; pseudonymized and aggregated

data can also be stored at the data owner area. (requirement a)

b. Data owner area is also divided to two separated areas: personalized data area and

impersonalized data area. This allows better to separate users who need personal information

from users who do not need them. (requirement a)

c. Access to the personal and pseudonymized data can only be done through secure

environments using highest-level authorization tools (corresponding to EIDAS “high”).

(requirement a)

d. Each dataset must be compiled using the different pseudonym for the same person.

(requirement a)

e. Along with data pseudonymizing, obfuscation is always applied. (requirement a)

f. “Aggregated data” is in most cases the synonym of “open data”; however, this is only

possible if a sufficient amount of data is aggregated on one line; the minimum number of

aggregated rows must be defined for each data set (requirement a)

g. It is mandatory to create isolated sandbox-like environments where researchers can analyse

(anonymized or aggregated) data; only aggregated (statistical) data can be extracted from

these environments. (requirement a)

h. All activities with the data must be stored in logs; it is necessary to build a log analysis

environment to detect misuse of data. (requirement b)

i. At the first stages of the project, it is necessary to find solutions for the creation of data

processing activities in accordance with current legislation; at the same time, it is necessary

to identify regulatory shortcomings that prevent data processing in the interests of

individuals and initiate changes to these laws. (requirement c)

j. The background, intent and goals of all partners must be verified every time when data usage

permission is granted; all data usage is monitored on the basis of log analysis; when the

misuse of data is detected, the misusers accounts will be closed until the decision is made

(requirement e)

APPENDIX 2 – Health Sense overall component architecture

APPENDIX 3 – Health Sense overall data flow chart