Possibilities and Utilization of Open Data in Healthcare Organisations
Henriksson, Kalle (2020-04-08)
Possibilities and Utilization of Open Data in Healthcare Organisations
(08.04.2020)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
suljettu
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe2020042722570
https://urn.fi/URN:NBN:fi-fe2020042722570
Tiivistelmä
Open data is a term used for data that is accessible for everyone to use, analyse or share without mentionable restrictions. Finnish public sector collects and maintains datasets on municipal-, region- and country-level, for example, about finance and healthcare. Usage and analysis of this data is scattered and usually done by public sector for reporting purposes using visual analytics and then quoted by private sector.
Focus for utilization of statistical methods, machine learning and artificial intelligence in healthcare tends to private persons’ medical data which has its own structural and juridical problems. The aim of this thesis is to research how publicly available datasets about Finnish healthcare could be used to allocate resources of healthcare organisations better and reduce their costs and is there restricting problems on open data for this.
The main challenge for utilizing open data is that it is maintained by different public sector organisations, meaning that data is published to various websites and file formats vary a lot. This problem is tackled in this thesis by pre-processing data to same format before analysis and systematically analysing data where finding of dependencies should be possible. In this thesis, Jupyter Notebook is used as the tool for data analysis and open data is tested and analysed both with statistical and machine learning methods.
The contribution of this thesis is an overall view about sources for open data in Finland and what kind of data they include and a case study which shows an example of how a healthcare organisation could analyse and benefit from publicly open data with analytical methods presented in this thesis. Case study successes in showing how costs can be predicted using data about the number of patients of certain diseases and how to measure dependencies of diseases and costs.
Focus for utilization of statistical methods, machine learning and artificial intelligence in healthcare tends to private persons’ medical data which has its own structural and juridical problems. The aim of this thesis is to research how publicly available datasets about Finnish healthcare could be used to allocate resources of healthcare organisations better and reduce their costs and is there restricting problems on open data for this.
The main challenge for utilizing open data is that it is maintained by different public sector organisations, meaning that data is published to various websites and file formats vary a lot. This problem is tackled in this thesis by pre-processing data to same format before analysis and systematically analysing data where finding of dependencies should be possible. In this thesis, Jupyter Notebook is used as the tool for data analysis and open data is tested and analysed both with statistical and machine learning methods.
The contribution of this thesis is an overall view about sources for open data in Finland and what kind of data they include and a case study which shows an example of how a healthcare organisation could analyse and benefit from publicly open data with analytical methods presented in this thesis. Case study successes in showing how costs can be predicted using data about the number of patients of certain diseases and how to measure dependencies of diseases and costs.