Inexpensive high quality fundus camera systems enable imaging of retina for vision related health management and diagnosis at large scale. A computer based analysis system can help establish the general baseline of normal conditions vs. anomalous ones, so that different classes of retinal conditions can be classified. Advanced applications, ranging from disease screening algorithms, aging vs. disease trend modeling and prediction, and content-based retrieval systems can be developed. In this dissertation, I propose an analytical framework for the modeling of retina blood vessels to capture their statistical properties, so that based on these properties one can develop blood vessel mapping algorithms with self-optimized parameters. Then, other image objects can be registered based on vascular topology modeling techniques. On the basis of these low level analytical models and algorithms, the third major element of this dissertation is a high level population statistics application, in which texture classification of macular patterns is correlated with vessel structures, which can also be used for retinal image retrieval. The analytical models have been implemented and tested based on various image sources. Some of the algorithms have been used for clinical tests. The major contributions of this dissertation are summarized as follows: (1) A concise, accurate feature representation of retinal blood vessel on retinal images by proposing two feature descriptors Sp and Ep derived from radial contrast transform. (2) A new statistical model of lognormal distribution, which captures the underlying physical property of the levels of generations of the vascular network on retinal images. (3) Fast and accurate detection algorithms for retinal objects, which include retinal blood vessel, macular-fovea area and optic disc, and (4) A novel population statistics based modeling technique for correlation analysis of blood vessels and other image objects that only exhibit subtle texture changes.

No description available.

Context. Diabetic retinopathy is the most common cause of new cases of legal blindness in people of working age. Early diagnosis is the key to slowing the progression of the disease, thus preventing blindness. Retinal fundus image is an important basis for judging these retinal diseases. With the development of technology, computer-aided diagnosis is widely used. Objectives. The thesis is to investigate whether there exist specific regions that could assist in better prediction of the retinopathy disease, it means to find the best region in fundus image that works the best in retinopathy classification with the use of computer vision and machine learning techniques. Methods. An experiment method was used as research methods. With image segmentation techniques, the fundus image is divided into regions to obtain the optic disc dataset, blood vessel dataset, and other regions (regions other than blood vessel and optic disk) dataset. These datasets and original fundus image dataset were tested on Random Forest (RF), Support Vector Machines (SVM) and Convolutional Neural Network (CNN) models, respectively. Results. It is found that the results on different models are inconsistent. As compared to the original fundus image, the blood vessel region exhibits the best performance on SVM model, the other regions perform best on RF model, while the original fundus image has higher prediction accuracy on CNN model. Conclusions. The other regions dataset has more predictive power than original fundus image dataset on RF and SVM models. On CNN model, extracting features from the fundus image does not significantly improve predictive performance as compared to the entire fundus image.

Automated fundus image analysis plays an important role in the computer aided diagnosis of ophthalmologic disorders. A lot of eye disorders, as well as cardiovascular disorders, are known to be related with retinal vasculature changes. Many studies has been done to explore these relationships. However, most of the studies are based on limited data obtained using manual or semi-automated methods due to the lack of automated techniques in the measurement and analysis of retinal vasculature. In this thesis, a fully automated retinal vessel width measurement technique is proposed. This novel method models the accurate vessel boundary delineation problem in two-dimension into an optimal surface segmentation problem in threedimension. Then the optimal surface segmentation problem is transformed into finding a minimum-cost closed set problem in a vertex-weighted geometric graph. The problem is modeled differently for straight vessel and for branch point because of the different conditions in straight vessel and in branch point. Furthermore, many of the retinal image analysis needs the location of the optic disc and fovea as a prerequisite information, for example, in the analysis of the relationship between vessel width and the distance to the optic disc. Hence, a simultaneous optic disc and fovea detection method is presented, which includes a two-step classification of three classes. The major contributions of this thesis include: 1) developing a fully automated vessel width measurement technique for retinal blood vessels, 2) developing a simultaneous optic disc and fovea detection method, 3) validating the methods using multiple datasets, and 4) applying the proposed methods in multiple retinal vasculature analysis studies.

No

Retina located in the back of the eye is not only a vital part of human sight, but also contains valuable information that can be used in biometric security applications, or for the diagnosis of certain diseases. In order to analyze this information from retinal images, its features of blood vessels, microaneurysms and the optic disc require extraction and detection respectively. We propose a method to extract vessels called MF-FDOG. MF-FDOG consists of using two filters, Matched Filter (MF) and the first-order derivative of Gaussian (FDOG). The vessel map is extracted by applying a threshold to the response of MF, which is adaptively adjusted by the mean response of FDOG. This method allows us to better distinguish vessel objects from non-vessel objects. Microaneurysm (MA) detection is accomplished with two proposed algorithms, Multi-scale Correlation Filtering (MSCF) and Dictionary Learning (DL) with Sparse Representation Classifier (SRC). MSCF is hierarchical in nature, consisting of two levels: coarse level microaneurysm candidate detection and fine level true microaneurysm detection. In the first level, all possible microaneurysm candidates are found while the second level extracts features from each candidate and compares them to a discrimination table for decision (MA or non-MA). In Dictionary Learning with Sparse Representation Classifier, MA and non-MA objects are extracted from images and used to learn two dictionaries, MA and non-MA. Sparse Representation Classifier is then applied to each MA candidate object detected beforehand, using the two dictionaries to determine class membership. The detection result is further improved by adding a class discrimination term into the Dictionary Learning model. This approach is known as Centralized Dictionary Learning (CDL) with Sparse Representation Classifier. The optic disc (OD) is an important anatomical feature in retinal images, and its detection is vital for developing automated screening programs. Currently, there is no algorithm designed to automatically detect the OD in fundus images captured from Asians, which are larger and have thicker vessels compared to Caucasians. We propose such a method to complement current algorithms using two steps: OD vessel candidate detection and OD vessel candidate matching. The proposed extraction/detection approaches are tested in medical applications, specifically the case study of detecting diabetic retinopathy (DR). DR is a complication of diabetes that damages the retina and can lead to blindness. There are four stages of DR and is a leading cause of sight loss in industrialized nations. Using MF-FDOG, blood vessels were extracted from DR images, while DR images fed into MSCF and Dictionary and Centralized Dictionary Learning with Sparse Representation Classifier produced good microaneurysm detection results. Using a new database consisting of only Asian DR patients, we successfully tested our OD detection method. As part of future work we intend to improve existing methods such as enhancing low contrast microaneurysms and better scale selection. In additional, we will extract other features from the retina, develop a generalized OD detection method, apply Dictionary Learning with Sparse Representation Classifier to vessel extraction, and use the new image database to carry out more experiments in medical applications.

Retina located in the back of the eye is not only a vital part of human sight, but also contains valuable information that can be used in biometric security applications, or for the diagnosis of certain diseases. In order to analyze this information from retinal images, its features of blood vessels, microaneurysms and the optic disc require extraction and detection respectively. We propose a method to extract vessels called MF-FDOG. MF-FDOG consists of using two filters, Matched Filter (MF) and the first-order derivative of Gaussian (FDOG). The vessel map is extracted by applying a threshold to the response of MF, which is adaptively adjusted by the mean response of FDOG. This method allows us to better distinguish vessel objects from non-vessel objects. Microaneurysm (MA) detection is accomplished with two proposed algorithms, Multi-scale Correlation Filtering (MSCF) and Dictionary Learning (DL) with Sparse Representation Classifier (SRC). MSCF is hierarchical in nature, consisting of two levels: coarse level microaneurysm candidate detection and fine level true microaneurysm detection. In the first level, all possible microaneurysm candidates are found while the second level extracts features from each candidate and compares them to a discrimination table for decision (MA or non-MA). In Dictionary Learning with Sparse Representation Classifier, MA and non-MA objects are extracted from images and used to learn two dictionaries, MA and non-MA. Sparse Representation Classifier is then applied to each MA candidate object detected beforehand, using the two dictionaries to determine class membership. The detection result is further improved by adding a class discrimination term into the Dictionary Learning model. This approach is known as Centralized Dictionary Learning (CDL) with Sparse Representation Classifier. The optic disc (OD) is an important anatomical feature in retinal images, and its detection is vital for developing automated screening programs. Currently, there is no algorithm designed to automatically detect the OD in fundus images captured from Asians, which are larger and have thicker vessels compared to Caucasians. We propose such a method to complement current algorithms using two steps: OD vessel candidate detection and OD vessel candidate matching. The proposed extraction/detection approaches are tested in medical applications, specifically the case study of detecting diabetic retinopathy (DR). DR is a complication of diabetes that damages the retina and can lead to blindness. There are four stages of DR and is a leading cause of sight loss in industrialized nations. Using MF-FDOG, blood vessels were extracted from DR images, while DR images fed into MSCF and Dictionary and Centralized Dictionary Learning with Sparse Representation Classifier produced good microaneurysm detection results. Using a new database consisting of only Asian DR patients, we successfully tested our OD detection method. As part of future work we intend to improve existing methods such as enhancing low contrast microaneurysms and better scale selection. In additional, we will extract other features from the retina, develop a generalized OD detection method, apply Dictionary Learning with Sparse Representation Classifier to vessel extraction, and use the new image database to carry out more experiments in medical applications.

Changes in retinal structure are related to different eye diseases. Various retinal imaging techniques, such as fundus imaging and optical coherence tomography (OCT) imaging modalities, have been developed for non-intrusive ophthalmology diagnoses according to the vasculature changes. However, it is time consuming or even impossible for ophthalmologists to manually label all the retinal structures from fundus images and OCT images. Therefore, computer aided diagnosis system for retinal imaging plays an important role in the assessment of ophthalmologic diseases and cardiovascular disorders. The aim of this PhD thesis is to develop segmentation methods to extract clinically useful information from these retinal images, which are acquired from different imaging modalities. In other words, we built the segmentation methods to extract important structures from both 2D fundus images and 3D OCT images. In the first part of my PhD project, two novel level set based methods were proposed for detecting the blood vessels and optic discs from fundus images. The first one integrates Chan-Vese's energy minimizing active contour method with the edge constraint term and Gaussian Mixture Model based term for blood vessels segmentation, while the second method combines the edge constraint term, the distance regularisation term and the shape-prior term for locating the optic disc. Both methods include the pre-processing stage, used for removing noise and enhancing the contrast between the object and the background. Three automated layer segmentation methods were built for segmenting intra-retinal layers from 3D OCT macular and optic nerve head images in the second part of my PhD project. The first two methods combine different methods according to the data characteristics. First, eight boundaries of the intra-retinal layers were detected from the 3D OCT macular images and the thickness maps of the seven layers were produced. Second, four boundaries of the intra-retinal layers were located from 3D optic nerve head images and the thickness maps of the Retinal Nerve Fiber Layer (RNFL) were plotted. Finally, the choroidal layer segmentation method based on the Level Set framework was designed, which embedded with the distance regularisation term, edge constraint term and Markov Random Field modelled region term. The thickness map of the choroidal layer was calculated and shown.

Context- Vision is one of the five elementary physiologial senses. Vision is enabled via the eye, a very delicate sense organ which is highly susceptible to damage which results in loss of vision. The damage comes in the form of injuries or diseases such as diabetic retinopathy and glaucoma. While it is not possible to predict accidents, predicting the onset of disease in the earliest stages is highly attainable. Owing to the leaps in imaging technology,it is also possible to provide near instant diagnosis by utilizing computer vision and image processing capabilities. Objectives- In this thesis, an algorithm is proposed and implemented to classify images of the retina into healthy or two classes of unhealthy images, i.e, diabetic retinopathy, and glaucoma thus aiding diagnosis. Additionally the algorithm is studied to investigate which image transformation is more feasible in implementation within the scope of this algorithm and which region of retina helps in accurate diagnosis. Methods- An experiment has been designed to facilitate the development of the algorithm. The algorithm is developed in such a way that it can accept all the values of a dataset concurrently and perform both the domain transforms independent of each other. Results- It is found that blood vessels help best in predicting disease associations, with the classifier giving an accuracy of 0.93 and a Cohen’s kappa score of 0.90. Frequency transformed images also presented a accuracy in prediction with 0.93 on blood vessel images and 0.87 on optic disk images. Conclusions- It is concluded that blood vessels from the fundus images after frequency transformation gives the highest accuracy for the algorithm developed when the algorithm is using a bag of visual words and an image category classifier model. Keywords-Image Processing, Machine Learning, Medical Imaging

Studies of the vascular tree in the retina have applications not only in the medical field but also biometrics. The mathematical description of the retinal vasculature could help facilitate and improve the diagnosis of certain diseases, their automatic localization or to accelerate the identification and verification of individuals. The aim is to design and develop an algorithm that will automatically approximate major retinal vessels by parabolic, linear and kvartic functions. The main part of this thesis is therefore devoted to this issue, including vascular segmentation using Gabor filters, morphological erosion, thresholding, skeletonization and the resulting optimization of the approximation model. The quality of the produced algorithm is discussed in the summary.