Search

Purpose. To determine whether a variational Bayesian independent component analysis mixture model (vB-ICA-mm), a form of unsupervised machine learning, can be used to identify and quantify areas of progression in standard automated perimetry fields. Methods. In an earlier study, it was shown that a model using vB-ICA-mm can separate normal fields from fields with six different patterns of visual field loss related to glaucomatous optic neuropathy (GON) along maximally independent axes. In the present study, an independent group of 191 patient eyes (66 with ocular hypertension (OHT), 12 with suspected glaucoma by field, 61 with suspected glaucoma by disc, and 52 with glaucoma) with five or more standard visual fields under observation for a mean of 6.24 ± 2.65 years and 8.11 ± 2.42 visual fields were evaluated with the vB-ICA-mm. In addition, eyes with progressive GON (PGON) were identified (n = 39). Each participant had a series of fields tested, with each field entered independently and placed along the axes of the previously developed model. This allowed change in one pattern of visual field defect (along one axis) to be assessed relative to results other areas of that same field (no change along other axes). Progression was based on a slope falling outside the 5th and the 95th percentile limits of all slopes, with at least two axes not showing such a deviation in a given individual’s series of fields. Fields were also scored using Advanced Glaucoma Intervention Study (AGIS) and the Early Manifest Glaucoma Treatment Trial (EMGT) criteria. Results. Thirty-two of 191 eyes progressed on vB-ICA-mm by this definition. Of the 32, 22 had field loss at baseline, 7 had only GON, 3 were OHTs and 12 were from the 39 eyes (31%) with PGON. The vB-ICA-mm identified a higher percentage of progressing eyes in each diagnostic category than did AGIS or and the EMGT. Conclusions. The vB-ICA-mm can quantitatively identify progression in eyes with glaucoma by evaluating change in one or more patterns of the visual field loss while other areas or patterns remain stable. This may enable each eye to contribute to the determination of whether change is caused by true progression or by variability.

PURPOSE. To classify healthy and glaucomatous eyes using relevance vector machine (RVM) and support vector machine (SVM) learning classifiers trained on retinal nerve fiber layer (RNFL) thickness measurements obtained by scanning laser polarimetry (SLP). METHODS. Seventy-two eyes of 72 healthy control subjects (average age 64.3 8.8 years, visual field mean deviation 0.71 1.2 dB) and 92 eyes of 92 patients with glaucoma (average age 66.9 8.9 years, visual field mean deviation 5.32 4.0 dB) were imaged with SLP with variable corneal compensation (GDx VCC; Laser Diagnostic Technologies, San Diego, CA). RVM and SVM learning classifiers were trained and tested on SLP-determined RNFL thickness measurements from 14 standard parameters and 64 sectors (approximately 5.6° each) obtained in the circumpapillary area under the instrument-defined measurement ellipse (total 78 parameters). Tenfold cross-validation was used to train and test RVM and SVM classifiers on unique subsets of the full 164-eye data set and areas under the receiver operating characteristic (AUROC) curve for the classification of eyes in the test set were generated. AUROC curve results from RVM and SVM were compared to those for 14 SLP software-generated global and regional RNFL thickness parameters. Also reported was the AUROC curve for the GDx VCC software-generated nerve fiber indicator (NFI). RESULTS. The AUROC curves for RVM and SVM were 0.90 and 0.91, respectively, and increased to 0.93 and 0.94 when the training sets were optimized with sequential forward and backward selection (resulting in reduced dimensional data sets). AUROC curves for optimized RVM and SVM were significantly larger than those for all individual SLP parameters. The AUROC curve for the NFI was 0.87. CONCLUSIONS. Results from RVM and SVM trained on SLP RNFL thickness measurements are similar and provide accurate classification of glaucomatous and healthy eyes. RVM may be preferable to SVM, because it provides a Bayesian-derived probability of glaucoma as an output. These results suggest that these machine learning classifiers show good potential for glaucoma diagnosis. (Invest Ophthalmol Vis Sci. 2005;46: 1322‐1329) DOI:10.1167/iovs.04-1122

To determine whether neural network techniques can improve differentiation between glaucomatous and nonglaucomatous eyes, using the optic disc topography parameters of the Heidelberg Retina Tomograph (HRT; Heidelberg Engineering, Heidelberg, Germany).With the HRT, one eye was imaged from each of 108 patients with glaucoma (defined as having repeatable visual field defects with standard automated perimetry) and 189 subjects without glaucoma (no visual field defects with healthy-appearing optic disc and retinal nerve fiber layer on clinical examination) and the optic nerve topography was defined by 17 global and 66 regional HRT parameters. With all the HRT parameters used as input, receiver operating characteristic (ROC) curves were generated for the classification of eyes, by three neural network techniques: linear and Gaussian support vector machines (SVM linear and SVM Gaussian, respectively) and a multilayer perceptron (MLP), as well as four previously proposed linear discriminant functions (LDFs) and one LDF developed on the current data with all HRT parameters used as input.The areas under the ROC curves for SVM linear and SVM Gaussian were 0.938 and 0.945, respectively; for MLP, 0.941; for the current LDF, 0.906; and for the best previously proposed LDF, 0.890. With the use of forward selection and backward elimination optimization techniques, the areas under the ROC curves for SVM Gaussian and the current LDF were increased to approximately 0.96.Trained neural networks, with global and regional HRT parameters used as input, improve on previously proposed HRT parameter-based LDFs for discriminating between glaucomatous and nonglaucomatous eyes. The performance of both neural networks and LDFs can be improved with optimization of the features in the input. Neural network analyses show promise for increasing diagnostic accuracy of tests for glaucoma.

To compare the ability of several machine learning classifiers to predict development of abnormal fields at follow-up in ocular hypertensive (OHT) eyes that had normal visual fields in baseline examination.The visual fields of 114 eyes of 114 patients with OHT with four or more visual field tests with standard automated perimetry over three or more years and for whom stereophotographs were available were assessed. The mean (+/-SD) number of visual field tests was 7.89 +/- 3.04. The mean number of years covered (+/-SD) was 5.92 +/- 2.34 (range, 2.81-11.77). Fields were classified as normal or abnormal based on Statpac-like methods (Humphrey Instruments, Dublin, CA) and by several machine learning classifiers. The machine learning classifiers were two types of support vector machine (SVM), a mixture of Gaussian (MoG) classifier, a constrained MoG, and a mixture of generalized Gaussian (MGG). Specificity was set to 96% for all classifiers, using data from 94 normal eyes evaluated longitudinally. Specificity cutoffs required confirmation of abnormality.Thirty-two percent (36/114) of the eyes converted to abnormal fields during follow-up based on the Statpac-like methods. All 36 were identified by at least one machine classifier. In nearly all cases, the machine learning classifiers predicted the confirmed abnormality, on average, 3.92 +/- 0.55 years earlier than traditional Statpac-like methods.Machine learning classifiers can learn complex patterns and trends in data and adapt to create a decision surface without the constraints imposed by statistical classifiers. This adaptation allowed the machine learning classifiers to identify abnormality in visual field converts much earlier than the traditional methods.

To determine which machine learning classifier learns best to interpret standard automated perimetry (SAP) and to compare the best of the machine classifiers with the global indices of STATPAC 2 and with experts in glaucoma.Multilayer perceptrons (MLP), support vector machines (SVM), mixture of Gaussian (MoG), and mixture of generalized Gaussian (MGG) classifiers were trained and tested by cross validation on the numerical plot of absolute sensitivity plus age of 189 normal eyes and 156 glaucomatous eyes, designated as such by the appearance of the optic nerve. The authors compared performance of these classifiers with the global indices of STATPAC, using the area under the ROC curve. Two human experts were judged against the machine classifiers and the global indices by plotting their sensitivity-specificity pairs.MoG had the greatest area under the ROC curve of the machine classifiers. Pattern SD (PSD) and corrected PSD (CPSD) had the largest areas under the curve of the global indices. MoG had significantly greater ROC area than PSD and CPSD. Human experts were not better at classifying visual fields than the machine classifiers or the global indices.MoG, using the entire visual field and age for input, interpreted SAP better than the global indices of STATPAC. Machine classifiers may augment the global indices of STATPAC.