Infrastructure
The Department of Computer and Systems Sciences has several infrastructures for research. Among other things, we have databases, GPUs and lab environments.
AI Compute Hub
The Department of Computer and Systems Sciences (DSV) maintains a growing and diverse Graphics Processing Unit (GPU) infrastructure. It supports a wide range of AI-related research and education.
The current environment includes several high-performance servers equipped with a total of 17 GPUs, including recent additions such as Mimer (delivered in 2023) featuring eight Nvidia A5500 cards, alongside other systems with Titan RTX and 2080Ti GPUs. These machines provide substantial computational capacity, backed by multi-core AMD processors, large memory configurations, and high-speed NVMe storage, to meet the demands of modern AI workloads.
The infrastructure is continuously monitored, maintained, and updated to ensure reliability and performance. It serves not only DSV’s AI research community, but also master’s students engaged in thesis projects and advanced courses in AI and Data Science, providing them with hands-on access to state-of-the-art computing resources.
Access to the infrastructure is available to DSV researchers as well as collaborators within Stockholm University. External researchers may also gain access through formal research collaborations or agreements, subject to resource availability and security guidelines.
Digital Forensics Lab
The Department of Computer and Systems Sciences (DSV) hosts a digital forensics lab, connected to the research subject of cybersecurity.
Digital forensics involves collecting and analysing digital traces, for example from chats on a suspect’s phone. Digital evidence must be produced and handled in a way that is legally sound.
The lab includes equipment related to the internet of things (IoT). It also has tools for examining mobile phones. PhD students and researchers with expertise in the field can use the lab to study IoT devices and other embedded computer systems in detail, in order to identify traces of usage.
Distributed Immersive Participation Lab
The Distributed Immersive Participation Lab is dedicated to advancing research, education, and industry collaboration in IoT, edge computing, distributed data processing, tiny machine learning, inclusive design, distributed edge intelligence and immersive participation. It is located to the Department of Computer and Systems Sciences (DSV), Campus Kista. The lab welcomes students, researchers, and industry professionals who would like to experiment, collaborate, and develop groundbreaking solutions that drive technological progress.
Extrality Lab
The Extrality Lab is a creative space available to researchers, students and partners. It is located to the Department of Computer and Systems Sciences (DSV).
The lab is dedicated to advancing knowledge in immersive systems, tangible interaction and embodied computing, as well as enabling novel digital humanities research approaches. It actively engages in creating workshops and courses on how to use state-of-the-art immersive technologies and prototyping tools.
The Extrality Lab is equipped with a full suite of XR devices and professional media production tools. It also has a fabrication workshop for physical prototyping including machines for soldering, 3D printing, laser cutting and sowing, and materials such as e-textiles and silicone.
The lab provides researchers with the means to conduct systematic experimentation on the design and evaluation of emerging interaction paradigms. The physical and human resources available serves as a scientific and technical framework for research projects that are carried out together with both academic and industrial partners.
Swedish Health Record Research Bank
The Swedish Health Record Research Bank (or “Health Bank” for short) is a unique research infrastructure containing large sets of electronic patient records, such as the Stockholm EPR (Electronic Patient Record) Corpus. The EPR Corpus stems from the TakeCare electronic patient records system that is used at the Karolinska University Hospital. It contains data from over 512 clinical units – more than two million patients – mainly from the years 2006 to 2014, but also partly until 2021.
All patient records are deidentified. This corpus contains both structured and unstructured information. The structured information includes patients’ age, gender, ICD-10 diagnosis codes, ATC medication codes, lab and blood values, as well as admission and discharge time and date.
The unstructured part of the data set contains clinical text. The whole corpus contains over 2.8 billion words or 17.8 GB text. The structured part encompasses 6 GB data.
See annotated data (pdf in Swedish) pdf, 274 kB.
Health Bank is used and has been used in a number of research projects carried out by the Natural Language Processing Research Group. The research is approved by the Regional Ethical Review Board in Stockholm (Regionala Etikprövningsnämnden i Stockholm) and the Swedish Ethical Review Authority (Etikprövningsmyndigheten) under various research plans.
Read about the Natural Language Processing Research Group
We have developed a number of clinical text mining tools based on the Stockholm EPR Corpus.
Explore our clinical text mining tools
Contact professor Hercules Dalianis, Director Health Bank
For reference to the Health Bank, please use:
Dalianis, H., A. Henriksson, M. Kvist, S. Velupillai and R. Weegar (2015).
HEALTH BANK – A Workbench for Data Science Applications in Healthcare (pdf).
Proceedings of the CAiSE-2015 Industry Track co-located with 27th Conference on Advanced Information Systems Engineering (CAiSE 2015), J. Krogstie, G. Juell-Skielse and V. Kabilan, (Eds.), Stockholm, Sweden, June 11, 2015, CEUR, Vol-1381, pp 1-18, urn:nbn:de:0074-1381-0.
Last updated: 2025-12-11
Source: Department of Computer and Systems Sciences, DSV