Fluorescence in situ hybridization (FISH) is a cytogenetic technique used to detect and localize the presence or absence of specific DNA sequences on chromosomes. FISH uses fluorescent probes, each tagged with a different fluorophore, that bind to specific parts of the chromosome. Through multi-band fluorescence microscopy the positions where the fluorescent probes bound to the chromosomes can be displayed so as to derive information of clinical relevance based on the presence and position of the fluorescent probes.
Accurate Evaluation of HER-2 Amplification in FISH Images
A sample application of this technique targets the measurement of the amplification of the HER-2 gene within the chromosomes, that constitutes a valuable indicator of invasive breast carcinomas. This requires the application to a tumor tissue sample of fluorescent probes that attach themselves to the HER-2 genes in a process called hybridization. These fluorescent probes carry a marker that emit light when the probes bind to the HER-2 genes, and this makes them visible as green spots under a fluorescent microscope. Similarly, a different probe, carrying a marker that makes it visible as a orange spot under a fluorescent microscope, is used to target the centromere 17 (CEP-17). Measuring the ratio of HER-2 over CEP-17 dots within each nucleus and then averaging this ratio for a representative number of cells allows estimation of HER-2 amplification.
In this research we present a system that supports accurate estimation of the ratio of HER-2 over CEP-17 dots in FISH images of breast tissue samples. Compared to previous work, the system incorporates a model to associate with each segmented nucleus a reliability score that estimates the confidence of the measure of the ratio of HER-2 over CEP-17 dots within the nucleus. This enables the computation of values of the ratio using only nuclei with high reliability scores so as to extract a measure of the amplification of HER-2 versus CEP-17 dots that better conforms to the evaluation of the pathologist compared to the ratio averaged over all available nuclei.
In many application scenarios digital images play a basic role and often it is important to assess if their content is realistic or has been manipulated to mislead watcher’s opinion. Image forensics tools provide answers to similar questions. We are working on a novel method that focuses in particular on the problem of detecting if a feigned image has been created by cloning an area of the image onto another zone to make a duplication or to cancel something awkward.
Image forensics using SIFT features
The proposed approach is based on SIFT features and allows both to understand if a copy-move attack has occurred and which are the image points involved, and, furthermore, to recover which has been the geometric transformation happened to perform cloning, by computing the transformation parameters. In fact when a copy-move attack takes place, usually an affine transformation is applied to the image patch selected to fit in a specified position according to that context. Our experimental results confirm that the technique is able to precisely individuate the attack and the transformation parameter estimation is highly reliable.
The aim of this project is to develop a new method of estimating the poses of imaged scene surfaces provided that they can be locally approximated by their tangent planes. Our approach performs an accurate direct estimation by exploiting the robustness of scale invariant feature transform (SIFT). The results are representative of the state of the art for this challenging task.
Local pose estimation from a single scale invariant keypoint
Retrieving the poses of keypoints in addition to matching them is an essential task in many computer-vision applications to transform uncostrained problems into costrained ones. This project proposes a new method of estimating the poses of regions around keypoints provided that they can be considered locally planar. While this has previously been attempted by adapting iterative algorithms developed for template matching, no explicit accurate direct estimation has been introduced before. Our approach simultaneously learn the “nuisance residual” structure present in the detection and description steps of the SIFT algorithm allowing local perspective properties of distinctive features to be recovered through a homography. The system is trained using synthetic images generated from a single reference view of the surface.
The method produces accurate detailed and fine grained set of local pose which can also be applied to non rigid surfaces. In particular the accuracy and robustness of the method are representative of the state of the art for this challenging task. At present, we investigate the application of the estimated homographies for building a pose-invariant descriptor for 3D face recognition.