IronDB: MR - based Analysis, Indexing, and Retrieval of Brain Iron Deposition in Basal Ganglia
Mujdat Cetin (Local Project Manager)
Aytul Ercil (Local Project Manager)
Ahmet Ekin (Philips Senior Scientist - Project Manager)
Octavian Soldea (Postdoc Sabanci University)
Devrim Unay (Philips)
Gokhan Uzunbas (alumni)
EU Marie Curie Program
The aging of the population in Europe necessitates more effective healthcare solutions. The discipline of neurology is seriously affected by this change in demographics, because advanced age is one of the greatest risk factors for certain neurological diseases, that impose significant financial and emotional costs to the society. The European Brain Council estimates the costs related to neurological diseases in Europe as 139 billion euros per year. Among others, this figure does not even include the early stages of the disease, the emotional pain of the family, and the loss of experienced people from the workforce. Early detection is essential to alleviate or even prevent these issues, but requires thorough understanding of the chemical changes in the brain.
One of the most salient chemical changes associated with neurodegenerative diseases, such as Alzheimer’s and Parkinson’s, is a large amount of iron accumulation in the basal ganglia (BG) organ of the brain. Because of this, it is often claimed that iron could be a potential biomarker for such diseases. This is very appealing because tissues with high iron concentration appear hypo-intense (dark) in the T2 contrast of MR images, which non-invasively provides varying intensity values as a function of chemical properties of tissues.
To exploit this idea, this project aims at
(1) determining automatic brain MR image analysis methods to detect iron in the BG, and
(2) developing search and retrieval tools to make it possible for the medical experts to interpret brain MR images by the similarity of iron accumulation in a large population.
The major benefits are threefold.
Firstly, in contrast to using invasive, costly, and potentially poisonous contrast agents for imaging and analysis of the brain at the molecular level, MR-based iron detection and analysis is a cost-effective, fast, and non-invasive diagnostic tool.
Secondly, in addition to the detection of iron in BG, we go one step further and propose search and retrieval tools to allow for comparisons of large number of patients to correlate the iron-related features with neurological diseases as well as other non-obvious pathologies, such as high blood pressure. This type of functionality does not exist today and it will be revolutionary.
As a third, computerized medical tools are highly desirable for other reasons as well. They speed up the time-consuming and error-prone manual tasks and improve the diagnostic quality by reducing the variability in human reasoning that may stem from mental and physical tiredness or experience level of the medical expert.
The project requires multi-disciplinary research involving the areas of image processing, pattern recognition and machine learning, databases, and clinical science. Therefore, Philips Research and Sabanci University jointly have initiated this project proposal. The capabilities of the two partners are complementary and critical for the success of the project.
Read the article about IronDB titled "Indexing and Retrieval of Brain MR biomarkers for diagnostic support":
Download the article:
In the framework of the "MR - based Analysis, Indexing, and Retrieval of Brain Iron Deposition in Basal Ganglia" project, the following papers are published:
 - Devrim Unay, "An Industry-Academia Partnership on Analysis, Indexing and Retrieval of MR Images to Unravel Brain Iron Deposition in Basal Ganglia," Proc. of EuroScience Open Forum – Marie Curie Action (ESOF-MC), Barcelona – Spain, 2008.