Big Data can help leukemia patients and prematurely born babies
D4Health-project tries to figure out how already existing data masses could be utilized efficiently in the treatment of different diseases. The D4Health – Data-Driven Decision Support for Digital Health Care project is funded by the Academy of Finland. It includes the Department of Computer Science and the Department of Electrical Engineering and Automation from the Aalto University, as well as the HIIT – a joint research institute of the University of Helsinki and the Aalto University - and the FIMMM – Institute for Molecular Medicine Finland.
The funding of the project is 1,2 million euros. It aims at developing methods for combining large medical datasets and prediction models, as well as creating flexible user interfaces that support physicians in diagnosing diseases and planning the treatment. Medical staff has an important role in this project.
– We collaborate closely with the HUS, states Professor Juho Rousu, who is in charge of the project.
– By merging digital medical records of a patient we try to describe the phenotype and then compare it to the genetic background of her/him. Our goal is to gain a more precise and personalized concept of the disease and in the best case even complete it with measurements and observations based on data-driven technologies, states Research Director Ari Lindqvist from the HUS.
New digital methods combine data from different sources and support physicians in making more efficient diagnoses.
Three pilot projects
Three pilots are included in the project: prematurely born babies, leukemia and Chronic obstructive pulmonary disease (COPD).
– In terms of prematurely born babies our goal is to predict the course of the disease from the very first sensitive signs and construct prediction models based on them. The ability to deduce fast and efficiently improves because of the artificial intelligence, explains Lindqvist.
Genomic measurements have been conducted of a few thousands of leukemia and chronic obstructive pulmonary disease patients. With the help of them it is possible to get information on how genes work, which mutations might occur and how drugs affect on gene functions.
– Certain types of gene variants may be more responsive to certain drugs, says Rousu.
– Computational network modelling platforms may enable even personalized drug treatments for severe cases of leukemia by using the patients' individual genome profiles, says Professor Tero Aittokallio, the FIMM group leader in charge of the leukemia pilot in collaboration of researchers from the FIMM, the HIIT and the HUS.
– BC Platforms Oy, a company specialized in genome data management, as well as Kustannus Oy Duodecim, participate in the project as industry partners, Rousu continues.
Standard data related to diseases is used for the development and testing of the models and the software. Privacy of the data is guaranteed by first removing the identification data. The analysis of the most sensitive data takes place in the HUS premises.
This research project uses machine learning and is in the crossroads of two key research areas of the Aalto University – digitalization as well as health and wellbeing. It is even related to the Health Capital Helsinki project, which aims at developing a Life Science research and company hub in the Helsinki metropolitan area.
Professor Juho Rousu
Tel: +358 50 415 1702
Aalto Health Platform: http://health.aalto.fi/en/