Huiying Liu

A survey on computer aided diagnosis for ocular diseases

BackgroundComputer Aided Diagnosis (CAD), which can automate the detection process for ocular diseases, has attracted extensive attention from clinicians and researchers alike. It not only alleviates the burden on the clinicians by providing objective opinion with valuable insights, but also offers early detection and easy access for patients.MethodWe review ocular CAD methodologies for various data types. For each data type, we investigate the databases and the algorithms to detect different ocular diseases. Their advantages and shortcomings are analyzed and discussed.ResultWe have studied three types of data (i.e., clinical, genetic and imaging) that have been commonly used in existing methods for CAD. The recent developments in methods used in CAD of ocular diseases (such as Diabetic Retinopathy, Glaucoma, Age-related Macular Degeneration and Pathological Myopia) are investigated and summarized comprehensively.ConclusionWhile CAD for ocular diseases has shown considerable progress over the past years, the clinical importance of fully automatic CAD systems which are able to embed clinical knowledge and integrate heterogeneous data sources still show great potential for future breakthrough.

Determining the difference in eyegaze measurements in individuals with Age related Macular Degeneration

Age-related Macular Degeneration (AMD) is one of the leading causes of blindness in the elderly. Visual loss associated with AMD often results in a central scotoma which is an alteration in the central vision, leading to distortion or loss of vision. Current methods of detecting AMD are typically manual, require holding fixation and an external response trigger. In this paper, we propose the use of eyegaze tracking to detect for the presence of AMD, using a simple set of test patterns. Experimental results show that the derived eyegaze measurements can help to identify individuals with AMD from healthy individuals. This could lead to the detection of AMD using eye tracking data, and could result in a potential system device for screening.Age-related Macular Degeneration (AMD) is one of the leading causes of blindness in the elderly. Visual loss associated with AMD often results in a central scotoma which is an alteration in the central vision, leading to distortion or loss of vision. Current methods of detecting AMD are typically manual, require holding fixation and an external response trigger. In this paper, we propose the use of eyegaze tracking to detect for the presence of AMD, using a simple set of test patterns. Experimental results show that the derived eyegaze measurements can help to identify individuals with AMD from healthy individuals. This could lead to the detection of AMD using eye tracking data, and could result in a potential system device for screening.

Local patch reconstruction framework for optic cup localization in glaucoma detection

Optic cup localization/segmentation has attracted much attention from medical imaging researchers, since it is the primary image component clinically used for identifying glaucoma, which is a leading cause of blindness. In this work, we present an optic cup localization framework based on local patch reconstruction, motivated by the great success achieved by reconstruction approaches in many computer vision applications recently. Two types of local patches, i.e. grids and superpixels are used to show the variety, generalization ability and robustness of the proposed framework. Tested on the ORIGA clinical dataset, which comprises of 325 fundus images from a population-based study, both implementations under the proposed frameworks achieved higher accuracy than the state-of-the-art techniques.

Effective Drusen Segmentation from Fundus Images for Age-Related Macular Degeneration Screening

Automatic screening of Age-related Macular Degeneration (AMD) is important for both patients and ophthalmologists. The major sign of contracting AMD at the early stage is the appearance of drusen, which are the accumulation of extracellular material and appear as yellow-white spots on the retina. In this paper, we propose an effective approach for drusen segmentation towards AMD screening. The major novelty of the proposed approach is that it employs an effective way to train a drusen classifier from a weakly labeled dataset, meaning only the existence of drusen is known but not the exact locations or boundaries. We achieve this by employing Multiple Instance Learning (MIL). Moreover, our proposed approach also tracks the drusen boundaries by using Growcut, with the output of MIL as initial seeds. Experiments on 350 fundus images with 96 of them with AMD demonstrates that our approach outperforms the state-of-the-art methods on the task of early AMD detection and achieves satisfying performance on the task of drusen segmentation.

Detecting AMD caused vision scotoma through eye tracking

Age Related Macular Degeneration (AMD) is the third leading cause of blindness and the first one in the elderly. AMD usually causes vision loss as the central vision field, e.g., vision scotoma, blur, and distortion, et. al. In this paper, we propose to detect AMD caused vision scotoma through eye tracking. Compared with the current vision assessment methods, e.g., Amsler grid, Microperimetry and Preferential Hyperacuity Perimetry, the proposed method has several advantages. 1) It does not require the patient to stare at a fixed position throughout the test. 2) It does not require the patient to orally or manually report / mark out the vision impairment. 3) It is easy to operate thus a trained nurse is capable of operating the test. To implement the proposed method, we collected gaze data of 138 eyes of 75 patients, who are diagnosed as AMD patient by clinicians. Nidek Microperimetry is adopted as the gold standard to get the ground truth and to evaluate our method. The result verifies the effectiveness of detecting vision scotoma through eye tracking.