Minister of science Helge Sander has just awarded the Free Research Council’s Young Elite Researcher prize to Michael Sass Hansen from DTU Informatics. The prize is awarded for his medical image analysis of the brain.
It is not only the minister of science who has an eye for talent. At Hvidovre Hospital a watchful eye has long been kept on Michael Sass Hansen’s research. Michael has just finished a PhD in medical image analysis. Speaking in layman’s terms he has developed a method which among other things is meant to help doctors ‘read’ MR images of brains. Ellen Garde, doctor in the department of ‘Aging and dementia’ at Hvidovre Hospital’s MR section explains:
“Over time we have become incredibly good at creating very detailed 3D MR images. The challenge consists of the image being very rich in information, making it very hard for the eye to identify errors or the hint of illnesses. With the help of Michael’s mathematical algorithms we hope that it will become easier and faster."
When brains become algorithms
Michael’s research work is based on 61 brain scans. The images derive from elderly people who are all in the same risk group of developing dementia related illnesses such as Alzheimer. Michael has concentrated his research specifically around corpus callosum which is the bridge that links the left and right halves of the brain. The goal has been to come up with a set of mathematical algorithms so that a computer can identify even tiny changes in the shape of corpus callosum quickly and flexibly. In time, this method may help determine if a patient suffers from dementia. By utilizing so called ‘Multi-Variable Splines’ Michael’s results have been ground-breaking:
“When you ask the computer to analyze the details of a scanned image, you previously used ‘uniform Splines’ that had a form of grid. In my research I utilized ‘Multi-Variable Splines’ that most of all resembles a cobweb. The benefit of using the ‘cobweb’ has turned out to be that the computer can handle the tiny statistical differences more easily while results are coming in faster at the same time,” says Michael Sass Hansen.
Michael Sass Hansen (Photo: Thorkild Christensen)
At Hvidovre Hospital’s MR section they are sincerely happy for Michael’s research results: They contribute to getting one step closer to answer the many unanswered questions regarding dementia:
“Unfortunately we doctors experience that the medicine that is available to patients with dementia today does not always have the desired effect. This is deeply frustrating for both patients and relatives. We have become more proficient at telling if people have dementia, but it is harder to identify exactly what type of dementia a given patient suffers from. It is here that new technologies such as MR scanning can help. Therefore it is completely invaluable that we have a person such as Michael to help us take the next step in analyzing the large quantity of data behind the images,”, says Ellen Garde.
Next stop: post.doc
While Doctor Ellen Garde continues to learn more about the brain and dementia, the next two years of Michael Sass Hansen’s life will be focused on the entire human body. With a prize of 200.000 DKR from the Free Research Council Michael Sass Hansen now journeys to Boston where he has just acquired a two year post.doc residency with Harvard Medical School. Here he will test and adjust his research in medical image analysis in a clinical environment:
“Today many operations are performed through the vein. During such operations it is necessary to take photos here and now of the procedure and compare them to the scanning image taken prior to the operation. The quality of the x-ray images available today is so poor that they are useless to the surgeon. Therefore, my goal during my post.doc is to develop methods so the surgeon can receive sharp, computer generated images during the operation. If this is successful it will ultimately mean faster, better and more secure operations,” says Michael Sass Hansen.
Rasmus Larsen, professor at DTU Informatics has been Michael Sass Hansen’s primary supervisor during his PhD project, and he has no doubts as to whether Michael returns home with good results from Harvard Medical School:
“Michael Sass Hansen is a highly talented researcher. That, combined with him spending two years in the Harvard research environment which is undoubtedly the largest and most powerful in the world in this area will be of tremendous value to Michael – and by extension, DTU,” says Rasmus Larsen.
Even though Michael Sass Hansen immigrates to Boston, the beneficial collaboration between Hvidovre Hospital and DTU Informatics continues. Rasmus Larsen recently initiated two new PhD projects, both focusing on medical image analysis.