Docent degrees

1. Title: Segmentering och klassificering av fluorescensmikroskopibilder - hur man undviker att snubbla på trösklar
   (Segmentation and classification in fluorescence microscopy images - how to avoid stumbling on thresholds)
   Carolina Wählby
   Date: 090122
   Abstract: Highly specific staining methods and fluorescent biological markers emitting light at different wavelengths together with fluorescence microscopy allow for detailed studies of the spatial distribution, localization, and interaction of biomolecules. Quantification of colocalization of biomolecules includes not only calculating a global measure of the degree of colocalization within an image, but also a classification of each image pixel showing colocalized signals or not. Common methods for automated quantification of colocalization require manual input of thresholds, and will give results strongly biased by the user. I present a novel, automated method for quantification of colocalization and classification of image pixels based on hue rather than intensity. The hue distribution is presented as an angle histogram created by a series of steps that compensate for the quantization noise always present in digital image data. Classification rules are thereafter based on the shape of the angle histogram, and detection of colocalized signals is thus only dependent on hue, making it possible to classify also low-intensity objects in noisy images, and decouple image segmentation from detection of colocalization. A number of examples from real biological applications are also presented.
   Comment: The docent lecture was held in Swedish.