Andrea Petznick

Data mining technique for automated diagnosis of glaucoma using higher order spectra and wavelet energy features

Eye images provide an insight into important parts of the visual system, and also indicate the health of the entire human body. Glaucoma is one of the most common causes of blindness. It is a disease in which fluid pressure in the eye increases gradually, damaging the optic nerve and causing vision loss. Robust mass screening may help to extend the symptom-free life for the affected patients. The retinal optic nerve fiber layer can be assessed using optical coherence tomography, scanning laser polarimetry (SLP), and Heidelberg Retina Tomography (HRT) scanning methods. These methods are expensive and hence a novel low cost automated glaucoma diagnosis system using digital fundus images is proposed. The paper discusses the system for the automated identification of normal and glaucoma classes using Higher Order Spectra (HOS) and Discrete Wavelet Transform (DWT) features. The extracted features are fed to the Support Vector Machine (SVM) classifier with linear, polynomial order 1, 2, 3 and Radial Basis Function (RBF) to select the best kernel function for automated decision making. In this work, SVM classifier with kernel function of polynomial order 2 was able to identify the glaucoma and normal images automatically with an accuracy of 95%, sensitivity and specificity of 93.33% and 96.67% respectively. Finally, we have proposed a novel integrated index called Glaucoma Risk Index (GRI) which is made up of HOS and DWT features, to diagnose the unknown class using a single feature. We hope that this GRI will aid clinicians to make a faster glaucoma diagnosis during the mass screening of normal/glaucoma images.

APPLICATION OF EMPIRICAL MODE DECOMPOSITION (EMD) FOR AUTOMATED DETECTION OF EPILEPSY USING EEG SIGNALS

Epilepsy is a global disease with considerable incidence due to recurrent unprovoked seizures. These seizures can be noninvasively diagnosed using electroencephalogram (EEG), a measure of neuronal electrical activity in brain recorded along scalp. EEG is highly nonlinear, nonstationary and non-Gaussian in nature. Nonlinear adaptive models such as empirical mode decomposition (EMD) provide intuitive understanding of information present in these signals. In this study a novel methodology is proposed to automatically classify EEG of normal, inter-ictal and ictal subjects using EMD decomposition. EEG decomposition using EMD yields few intrinsic mode functions (IMF), which are amplitude and frequency modulated (AM and FM) waves. Hilbert transform of these IMF provides AM and FM frequencies. Features such as spectral peaks, spectral entropy and spectral energy in each IMF are extracted and fed to decision tree classifier for automated diagnosis. In this work, we have compared the performance of classification using two types of decision trees (i) classification and regression tree (CART) and (ii) C4.5. We have obtained the highest average accuracy of 95.33%, average sensitivity of 98%, and average specificity of 97% using C4.5 decision tree classifier. The developed methodology is ready for clinical validation on large databases and can be deployed for mass screening.

Extensive characterization of human tear fluid collected using different techniques unravels the presence of novel lipid amphiphiles

The tear film covers the anterior eye and the precise balance of its various constituting components is critical for maintaining ocular health. The composition of the tear film amphiphilic lipid sublayer, in particular, has largely remained a matter of contention due to the limiting concentrations of these lipid amphiphiles in tears that render their detection and accurate quantitation tedious. Using systematic and sensitive lipidomic approaches, we validated different tear collection techniques and report the most comprehensive human tear lipidome to date; comprising more than 600 lipid species from 17 major lipid classes. Our study confers novel insights to the compositional details of the existent tear film model, in particular the disputable amphiphilic lipid sublayer constituents, by demonstrating the presence of cholesteryl sulfate, O-acyl-ω-hydroxyfatty acids, and various sphingolipids and phospholipids in tears. The discovery and quantitation of the relative abundance of various tear lipid amphiphiles reported herein are expected to have a profound impact on the current understanding of the existent human tear film model.

Nuclear Factor-κB: Central Regulator in Ocular Surface Inflammation and Diseases

The nuclear factor-κB (NF-κB) is a key transcription factor pathway that is responsible for many key biological processes, such as inflammation, apoptosis, stress response, corneal wound healing, angiogenesis, and lymphangiogenesis. Numerous recent studies have investigated NF-κB in the context of ocular surface disorders, including chemical injury, ultraviolet radiation-induced injury, microbial infections, allergic eye diseases, dry eye, pterygium, and corneal graft rejection. The purpose this article is to summarize key findings with regard to the pathways regulating NF-κB and processes governed by the NF-κB pathway. In the innate defense system, NF-κB is involved in signaling from the toll-like receptors 2, 3, 4, 5 and 7, which are expressed in conjunctival, limbal, and corneal epithelial cells. These determine the ocular responses to infections, such as those caused by Pseudomonas aeruginosa, Staphylococcus aureus, adenovirus, and herpes simplex-1 virus. Natural angiogenic inhibitors enhance NF-κB, and this may occur through the mitogen-activated protein kinases and peroxisome proliferator-activated receptor γ. In alkali injury, inhibition of NF-κB can reduce corneal angiogenesis, suggesting a possible therapeutic strategy. The evaluation of NF-κB inhibitors in diseases is also discussed, including emodin, besifloxacin, BOL-303242-X (mapracorat), thymosin-β4, epigallocatechin gallate, Perilla frutescens leaf extract and IKKβ-targeting short interfering RNA.

APPLICATION OF INTRINSIC TIME-SCALE DECOMPOSITION (ITD) TO EEG SIGNALS FOR AUTOMATED SEIZURE PREDICTION

Intrinsic time-scale decomposition (ITD) is a new nonlinear method of time-frequency representation which can decipher the minute changes in the nonlinear EEG signals. In this work, we have automatically classified normal, interictal and ictal EEG signals using the features derived from the ITD representation. The energy, fractal dimension and sample entropy features computed on ITD representation coupled with decision tree classifier has yielded an average classification accuracy of 95.67%, sensitivity and specificity of 99% and 99.5%, respectively using 10-fold cross validation scheme. With application of the nonlinear ITD representation, along with conceptual advancement and improvement of the accuracy, the developed system is clinically ready for mass screening in resource constrained and emerging economy scenarios.

Lipidomic analysis of human tear fluid reveals structure-specific lipid alterations in dry eye syndrome

As current diagnostic markers for dry eye syndrome (DES) are lacking in both sensitivity and specificity, a pressing concern exists to develop activity markers that closely align with the principal axes of disease progression. In this study, a comprehensive lipidomic platform designated for analysis of the human tear lipidome was employed to characterize changes in tear lipid compositions from a cohort of 93 subjects of different clinical subgroups classified based on the presence of dry eye symptoms and signs. Positive correlations were observed between the tear levels of cholesteryl sulfates and glycosphingolipids with physiological secretion of tears, which indicated the possible lacrimal (instead of meibomian) origin of these lipids. Notably, we found wax esters of low molecular masses and those containing saturated fatty acyl moieties were specifically reduced with disease and significantly correlated with various DES clinical parameters such as ocular surface disease index, tear breakup time, and Schirmers I test (i.e., both symptoms and signs). These structure-specific changes in tear components with DES could potentially serve as unifying indicators of disease symptoms and signs. In addition, the structurally-specific aberrations in tear lipids reported here were found in patients with or without aqueous deficiency, suggesting a common pathology for both DES subtypes.

Longitudinal changes in tear fluid lipidome brought about by eyelid-warming treatment in a cohort of meibomian gland dysfunction.

Meibomian gland dysfunction (MGD) is a leading cause of evaporative dry eye and ocular discomfort characterized by an unstable tear film principally attributed to afflicted delivery of lipids to the ocular surface. Herein, we elucidated longitudinal tear lipid alterations associated with disease alleviation and symptom improvement in a cohort of MGD patients undergoing eyelid-warming treatment for 12 weeks. Remarkably, eyelid-warming resulted in stark reductions in lysophospholipids (P < 0.001 for lyso-plasmalogen phosphatidylethanolamine, lysophosphatidylcholine, and lysophosphatidylinositol), as well as numerous PUFA-containing diacylglyceride species in tears, accompanied by significant increases in several PUFA-containing phospholipids. These changes in tear lipidomes suggest that eyelid-warming leads to diminished activity of tear phospholipases that preferentially target PUFA-containing phospholipids. In addition, treatment led to appreciable increases (P < 0.001) in O-acyl-ω-hydroxy-FAs (OAHFAs), which are lipid amphiphiles critical to the maintenance of tear film stability. Longitudinal changes in the tear lipids aforementioned also significantly (P < 0.05) correlated with reduced rate of ocular evaporation and improvement in ocular symptoms. The foregoing data thus indicate that excess ocular surface phospholipase activity detrimental to tear film stability could be alleviated by eyelid warming alone without application of steroids and identify tear OAHFAs as suitable markers to monitor treatment response in MGD.

Choice of artificial tear formulation for patients with dry eye: where do we start?

Abstract: Dry eye is prevalent in many parts of the world. As a result, ophthalmologists and other health care professionals, such as optometrists and general practitioners, often help these patients manage their symptoms. The most common form of treatment for management of dry eye is over-the-counter tear lubricants. A number of tear lubricant formulations are available that vary by their mechanism of action. This article suggests simple guidelines on how lubricants can be selected for patients with mild to moderate dry eye. Side effects of lubricants, such as burning on instillation because of mismatches of eye drop with tear acidity, are also discussed.

Associations of systemic diseases, smoking and contact lens wear with severity of dry eye

Citation information: Lee S‐Y, Petznick A & Tong L. Associations of systemic diseases, smoking and contact lens wear with severity of dry eye. Ophthalmic Physiol Opt 2012, 32, 518–526. doi: 10.1111/j.1475‐1313.2012.00931.x

Detection of meibomian glands and classification of meibography images.

Computational methods are presented that can automatically detect the length and width of meibomian glands imaged by infrared meibography without requiring any input from the user. The images are then automatically classified. The length of the glands are detected by first normalizing the pixel intensity, extracting stationary points, and then applying morphological operations. Gland widths are detected using scale invariant feature transform and analyzed using Shannon entropy. Features based on the gland lengths and widths are then used to train a linear classifier to accurately differentiate between healthy (specificity 96.1%) and unhealthy (sensitivity 97.9%) meibography images. The user-free computational method is fast, does not suffer from inter-observer variability, and can be useful in clinical studies where large number of images needs to be analyzed efficiently.