Helle Rootzén

Helle Rootzén


Richard Petersens Plads

Building: 324, 253

2800 Kgs. Lyngby


Phone: 45259677

Mobile: 51647476

Professor at the Technical University of Denmark and from 2010 to 2015 director of DTU Compute. Is working on learning objects, e-learning, learning platforms and student based learning. Likes to look into the exponential future, leadership, statistics, big data, diversity - and hopes for more women designing the society.

Boards and thinktanks:

- Nordisk Kollegium, board (2006-10) 

- It-tænkerne, thinktank appointed by Dansk IT (2010- ) 

- DeIC (Danish e-Infrastructure Coorperation), board (2012- ) 

- Birkerød Gymnasium, HF, IB, and boarding school, board (2012- ) 

- The Danish Government appointed “Vækstteam for IKT and digital vækst”, thinktank 

- Dansk IT, board (2012-2013)

- DADIU (The National Academy of Digital Interactive Entertainment), board (2010-2015) 

- DEFF (Danmarks elektroniske fag og forskningsbibliotek), advisory board (2014- ) 

- ATV’s (Danish Academy of Technical Sciences)  “Digitale Vismænd”, thinktank (2014- ) 


Academic grades

Cand. Scient., Ph.d.


Student based learning in multi-facetted learning communities – STUDeNT-learn 

The project aims to explore, propose and test methods to promote student-driven e-learning. How do you create a similar "spirit" as experienced in hackerthons? Can you imagine that students get together for a "theme" and even organize the learning proces themselves - perhaps across competence levels? What is the role of the teachers? Will the weak students suffer - or will it further strengthen them? There are many questions and this project will try to answer some of them and will also identify whether there is a basis to describe a more extensive project that can be supported by other funding. A "side effect" of student driven / student based e-learning is to minimize resurses which is becoming more and more urgent in university teaching. When the number of students increases, there are requirements for propulsion, and an increased student demand for individualized instruction. It is also for this reason need to think in new ways.

Statistical/mathematical e-learning: HEROS 

I have developed and implemented a new approach to e-learning called HEROS (Higher Education 
Re-usable Objects in Statistics): a non-linear learning object based learning system. It lets students 
design their own individual research based course in statistics and mathematics in a way which fits 
their individual levels and learning styles. I have lead a DKK 3 million project funded by the 
Ministry of Science, Technology and Innovation to develop a continuing education course in 
statistics for industry and a university course in mathematics (e*math) – both based on the ideas in 
HEROS. One result is the current introductory course in mathematics at the Technical University of 
Denmark. The course is taken by around 800 students each year and is receiving very good student 
evaluations. The course is also adopted by Roskilde University and the University of the Faroe 
Islands, and has also been asked for by other universities, but as a bachelor course at DTU it is in 

Industrial Statistics: improving industrial processes  

The explosive increase of size and sophistication of modern industry leads to a need for 
development of statistical methods for improving industrial processes to obtain higher quality, 
reliability and productivity. However, the distance between theoretical research at universities and 
the practical problems facing industry is often large. The area is full of ad-hoc methods and new 
useful research results often find it difficult get used by industry. My goal is to bridge the gap 
between the two worlds, by combining development of new theory with work on concrete industrial 
projects. One example is production planning and design for the NovoPen4 at Novo Nordisk, as a 
pilot project for on-line statistical process control in Novo Nordisk (joint with Gitte Legindgaard 
and others). 
Environmental Statistic: quantifying phytoplankton response to changing nutrient levels 

Euthrification in coastal waters causes growth of phytoplankton which can lead oxygen depletion. 
Experimental studies and theoretical considerations of growth strategies have shown that changing 
nutrient levels should affect the phytoplankton composition. However, no real case studies had been 
done to support these theories. Using data spanning two decades from 30 stations on the coast of 
Denmark, we have created statistical models which quantify the phytoplankton response to total 
nitrogen and total phosphorous and which identifies species with high sensitivity to changing 
nutrient levels (joint with Hua Fan and Jacob Carstensen). 
Medical Statistics: prediction of breast cancer survival 

We studied the impact of lymph node status at diagnosis on survival of breast cancer patients. Our 
study included breast cancer patients diagnosed from 1978 to 1994 in Denmark, where the 
treatment schemes for breast cancer patients were fairly stable, and where mammography screening 
was limited. Changes in lymph node status distribution explained half of the improvement in 5-year 
relative survival, and seems to be the single most important cause of the improving survival of reast cancer patients in Denmark (joint with the Danish Breast Cancer Cooperation Group and others). 
Chemical statistics: calibration problems with many variables 

A basic problem in chemical and environmental engineering is calibration of advanced and high 
dimensional measuring systems. We compared different kinds of statistical methods proposed in the 
literature as to their suitability for solving calibration problems with many variables. The methods 
incljude principal component regression, partial least-squares, and ridge regression. It was found 
that the statistical techniques themselves did not provide robust results in the spirit of calibration 
equations which can last for long periods. Instead we showed how this can be acheived  by 
complementing the methods with smoothing and differention techniques (joint with DTU 
Telecommunication stochastics: stability of queuing systems 

Inspired by queuing problems in telecommunication we studied stability and instability of queuing 
systems, and were able to obtain concrete conditions which differentiated between these behaviours 
for queues with general arrival and server distributions, and for any number of servers (joint with 
Søren Asmussen). 


My files

View graph of relations

View all »

  • Acta Oncologica

    ISSNs: 0284-186X, 0284186x

    ISSNs (Electronic): 1100-170, 1100-1704

    Informa Healthcare, United Kingdom

    BFI (2015): BFI-level 1, ISI indexed (2013): ISI indexed yes

    Central database


  • Journal of Agricultural, Biological, and Environmental Statistics

    ISSNs: 1085-7117

    ISSNs (Electronic): 1537-2693

    American Statistical Association, United States

    BFI (2015): BFI-level 1, ISI indexed (2013): ISI indexed yes

    Central database


  • Politiken

    ISSNs: 0907-1814

    JP/Politikens Hus A/S, Denmark

    ISI indexed (2013): ISI indexed no

    Central database


View all »

Download as:
Download as PDF
Select render style:
Download as HTML
Select render style:
Download as Word
Select render style:
Download as: To download, please limit the number of search results to 1,000 or less.

Place of employment

Latest activities and conferences

ID: 2201685