Where Data Meets Computer Science
Header

MSc Projects

BioImage Informatics: 2D/3D Blob Detector

Project Type/Degree: CS / SE
Description: In the area of image processing, blob detection refers to algorithms that are aimed at detecting points or regions in the image that differ in properties like brightness or color compared to the surrounding. The student(s) doing this project will implement most common 2D/3D blob detector based on the Laplacian of the Gaussian (LoG). To obtain a multi-scale blob detector with automatic scale selection, the scale-normalized Laplacian operator will be considered. The performance of the implemented blob detector will be evaluated by its applications to blob-like objects recognition and tracking in 2D/3D biomedical images.
Requirements: Matlab, Java.
Keywords: image analysis and processing, blob-like features detection.


 

BioImage Informatics: 2D Coupled Active Contours

Project Type/Degree: CS / SE
Description: Active contours provide a very effective framework for image segmentation and object tracking. The student(s) doing this project will implement an extension of active contours designed to track non-occluding objects transiently touching each other. This technique minimizes a cost functional that depends on all contours simultaneously and includes a penalty for contour overlaps. The performance of the implemented coupled active contours approach will be evaluated by its applications to segmentation and tracking of touching objects in 2D biomedical images.
Requirements:  Matlab, Java.
Keywords: image analysis and processing, active contours, topology, segmentation.


 

BioImage Informatics: 2D Junction Detection Using Streamlines

Project Type/Degree: CS / SE
Description: The student(s) doing this project will implement an image processing method to detect multi-modal regions composed of linear structures and measure the orientations in these regions, i.e. at line X-sings, T-junctions and Y-forks. The method is based on streamlines which are constructed from a vector field that represents the local structure in the image. This method allows us to accurately locate junctions and crossings and at the same time measures the attributes of the underlying uni-modal structures. The performance of the implemented junction detection approach will be evaluated by its applications to branching points extraction in 2D biomedical networks.
Requirements: Matlab, Java.
Keywords: image analysis and processing, junction detection, streamlines, vector field.


 

BioImage Informatics: 3D Morphological Skeletonization

Project Type/Degree: CS / SE
Description: In the processing and analysis of images it is important to be able to extract features, describe shapes and recognize patterns. Such tasks refer to geometrical concepts such as size, shape, and orientation. Mathematical morphology uses concepts from set theory, geometry and topology to analyse geometrical structures in an image. In particular, morphological skeleton is a skeleton (or medial axis) representation of a shape or binary image, computed by means of morphological operators. The student(s) doing this project will implement a morphological skeletonization algorithm based on sets of structuring elements for hit-or-miss transforms whereas each structuring element actually describes a shape primitive. The performance of the implemented coupled active contours approach will be evaluated by its applications to centreline extraction in 3D biomedical networks.
Requirements: Matlab, Java.
Keywords: image analysis and processing, mathematical morphology, skeletonization.


 

BioImage Informatics: Mobility and Behaviour as Surrogate Measures of Pain

Project Type/Degree: CS / SE
Description: Abnormal posture and mechanics have long been associated with pain or injuries. Therefore, the measurement and classification of patient posture in a clinical setting has become a central focus of lower extremity medicine, and now is widely used to evaluate injury risk and monitor treatment efficacy and impact of various therapies. However, despite the existence of many different techniques to evaluate patient’s posture in the clinical setting, there is still disagreement as to which method is the most clinically useful.
To address these concerns, the student(s) will develop and validate a system able to provide a quantitative characteristic of static and dynamic aspects of the patient posture in clinical as well as home settings for everyday patient assessment. Consequently, such a system needs to be inexpensive, portable and accurate. In this respect, we will focus on a use of the Microsoft Kinect, which is an inexpensive and portable video game accessory that combines a video and infrared-sensing camera to create a 3D model of the body. The performance of the proposed approach will be evaluated in the clinical settings at The James Cook University Hospital in Middlesbrough.
Requirements: Matlab, Java.
Keywords: image analysis and processing, computer vision.


 

BioImage Informatics: Volumetric Assessment of Diabetic Macular Oedema

Project Type/Degree: CS / SE
Description:Diabetic retinopathy, one of the most frequent complications of diabetes, remains a major public health problem with significant socio-economic implications. It affects approximately 50% of diabetic subjects and remains the leading cause of blindness in working-age populations of industrialized countries. Diabetic macular oedema (DMO) is the most prevalent cause of visual acuity loss in patients with diabetes. Recent advances in OCT technology with the advent of spectral domain OCT (SD-OCT), has allowed a significantly higher axial image resolution. Volumetric analysis of 3D OCT images of eyes with DMO can provide a detailed analysis of various features that have been shown to be useful indicators of VA change and can potentially predict treatment response.
The objectives of this project are to develop and assess novel image analysis techniques of 3D OCT scans for volumetric assessment of different morphological patterns in DMO. The student(s) doing this project will utilise OCT scan en-face images extracted from the diabetic retinal imaging database at Sunderland Eye Infirmary captured by a Heidelberg Spectralis SD-OCT machine for 50 patients with DMO.
Requirements: Matlab, Java.
Keywords: image analysis and processing, visualisation.

For further information please contact Dr Boguslaw Obara (boguslaw [dot] obara [AT] durham [dot] ac [dot] uk)