Nucleotyping

The methods are based on computer assisted image analysis, where we transfer images of cell and tissue samples from microscope (light, laser scan and electron microscopy) to the computer. In the computer we can digitalise the images and analyse the heritage materials structure and organisation bit by bi. These kinds of examinations of interphase nuclei are called nucleotyping.

Nucleotyping are among other an objective measure of nuclei atypi, one of the most important parameters for pathological diagnostics and prognosis. A qualitative analysis of chomatin structure is measured simultaneously with a measure of total amount of chromatin or DNA (depending on staining method) in nuclei. The method give information on the DNA ploidy distribution in addition to traditional nuclei morphometry, and it is sensitive to larger chormosomal aberrations. Of even greater importance is the methods ability to map and quantify functional changes in DNA organisation, that may be induced of larger or smaller mutations. Such changes is to al large ectent subvisual, and will not be discovered using ordinary microscopy. Nucletyping can be described as interphase cytogenetics, that is an interphase variant of chomosomal analysis where we map and describe organisatory and functionally domenes of DNA.

Nucleotyping has a great potential as diagnostic and prognostic/predictive marker in cancer. In opposite to DNA ploidy this method has also a predictive value in cancer at advanced stages.

Nucleotyping is a main priority area at Imaging programme, and we have a number of large studies going on prostate cancer, breast cancer (DCIS and Stage I) and colon cancer, both at The Norwegian Radium Hospital and at institutions in England.


Segmentation

Segmentation is one of the most important and difficult steps in automated image analysis.
In segmentation of cell nuclei the difficulty highly depends on the type of specimen to be analyzed.
Nuclei in cytological specimens may be segmented by simple automatic gray level thresholding.
Nuclei in tissue sections, however, is in general very difficult to segment by purely automatic means.
The aim of the project is to develop automatic methods for segmentation of nuclei in histological sections.
 
Mar 9, 2007 Page visits: 4248