Optical Coherence Tomography Machine Learning Classifiers for Glaucoma Detection: A Preliminary Study

Burgansky-Eliash, Zvia; Wollstein, Gadi; Chu, Tianjiao; Ramsey, Joseph D.; Glymour, Clark; Noecker, Robert J.; Ishikawa, Hiroshi; Schuman, Joel S.

doi:10.1167/iovs.05-0366

Cited by 178 publications

(126 citation statements)

References 34 publications

Supporting

Mentioning

109

Contrasting

Unclassified

Order By: Relevance

“…MLC techniques have also been employed with various technologies designed to perform structural and functional evaluation of glaucoma, including TD-OCT (15,17,19,23) , HRT (18,22,24) , GDx (16,20) , and visual field (13,18,20,21) . Several studies have used MLC techniques to combine functional and imaging data in attempt to improve diagnostic accuracy (19,22) .…”

Section: Discussionmentioning

confidence: 99%

“…Studies comparing these two technologies have demonstrated that the sensitivity and specificity of various RNFL parameters using the Cirrus OCT are excellent and equivalent to the Stratus OCT (9)(10)(11)(12)(13) . Since 1990 (Goldbaum MH, et al IOVS 1990;31; ARVO Abstract 503), machine learning classifier (MLC) techniques have been applied to optical imaging and visual function measurements to improve glaucoma detection, with results suggesting that these techniques are as good as or better than currently available methods at classifying eyes as glaucomatous or healthy (14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24) . Classifiers usually employ a form of supervised learning, where the program learns from positive and ne gative training examples, representing cases where, for example, there are signs of glaucoma on data obtained by examination of the visual field (positive examples) or not (negative examples).…”

Section: Sensitivity and Specificity Of Machine Learning Classifiers mentioning

confidence: 99%

See 1 more Smart Citation

Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry

Silva¹,

Vidotti²,

Cremasco³

et al. 2013

Arq. Bras. Oftalmol.

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Sensitivity and Specificity Of Machine Learning Classifiers mentioning

confidence: 99%

Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry

Silva¹,

Vidotti²,

Cremasco³

et al. 2013

Arq. Bras. Oftalmol.

View full text Add to dashboard Cite

“…Measurements of the circumpapillary RNFL are reproducible [7][8][9] and useful in the early diagnosis of glaucoma. [10][11][12][13][14][15][16][17][18][19] Stratus (Carl Zeiss Meditec, Inc., Dublin, CA) is the most widely used OCT system for glaucoma diagnosis. Stratus measures the RNFL thickness profile along a 3.4-mm-diameter circle around the optic disc.…”

mentioning

confidence: 99%

Combining Nerve Fiber Layer Parameters to Optimize Glaucoma Diagnosis with Optical Coherence Tomography

Lu¹,

Wang²,

Varma³

et al. 2008

Ophthalmology

Self Cite

View full text Add to dashboard Cite

Purpose-To identify the best combination of Stratus optical coherence tomography (OCT) retinal nerve fiber layer (RNFL) thickness parameters for the detection of glaucoma. Design-Observational cross-sectional study.Participants-Eighty-nine age-matched normal and perimetric glaucoma participants enrolled in the Advanced Imaging for Glaucoma Study.Methods-The Zeiss Stratus OCT system was used to obtain the circumpapillary RNFL thickness in both eyes of each participant. Right and left eye clock-hour data are analyzed together, assuming mirror-image symmetry. The RNFL diagnostic parameters were combined using either or-logic or and-logic approaches.Main Outcome Measures-Area under the receiver operating characteristic curve (AROC), sensitivity, and specificity are used to evaluate diagnostic performance.Results-Overall average RNFL thickness has the highest AROC value (0.89) of all single parameters evaluated, followed by the inferior and superior quadrants (0.88 and 0.86, respectively). The clock hours with the best AROC values are in the inferior and superior quadrants. The highest AROC (0.92) was achieved by the or-logic combination of overall, inferior, and superior quadrant RNFL thicknesses. The 3-parameter combination was significantly better than the overall average alone (P = 0.01) The addition of more quadrants or clock hours to the combination reduced diagnostic performance.Conclusions-The best stand-alone diagnostic strategy for Stratus OCT RNFL data is to classify an eye as glaucomatous if the overall, inferior quadrant, or superior quadrant RNFL thickness average is below normal.Glaucoma is an optic neuropathy characterized by irreversible loss of neural tissue and visual field (VF). Currently, definitive glaucoma diagnosis is based on VF testing. However, between 30% and 50% of the ganglion cells may be lost before abnormalities appear in perimetric Optical coherence tomography, first described by Huang et al, 4 is the only imaging modality with sufficiently fine depth resolution to measure the RNFL thickness directly. Measurements of the circumpapillary RNFL are reproducible 7-9 and useful in the early diagnosis of glaucoma. [10][11][12][13][14][15][16][17][18][19] Stratus (Carl Zeiss Meditec, Inc., Dublin, CA) is the most widely used OCT system for glaucoma diagnosis. Stratus measures the RNFL thickness profile along a 3.4-mm-diameter circle around the optic disc. Based on the profile, the averages of overall, quadrant, and clockhour RNFL thickness are computed and displayed. At present, the clinician has no clear guideline on whether one, several, or all of these diagnostic parameters should be used in the clinical diagnosis of glaucoma. Based on the area under the receiver operating characteristic curve (AROC), some studies have shown the overall RNFL thickness average to be the best diagnostic parameter. 11,13,14,16,19 Other studies have shown the inferior 10,17,20 or superior 21 quadrant RNFL thickness average to be the best, in agreement with clinical observation that glaucomatous optic...

show abstract

“…SVMs and other machine learning classifiers have previously been used for various classification problems, including automatic and semiautomatic retinal layer segmentation by classifying pixels as belonging to different layers [21,22], glaucoma detection [23][24][25][26][27], and segmentation of the ONH [28]. In this work, we exploit a novel utilization of SVMs to detect the images of diseased eyes in which some retinal layers may be missing.…”

Section: Svmmentioning

confidence: 99%

Automatic segmentation of up to ten layer boundaries in SD-OCT images of the mouse retina with and without missing layers due to pathology

Srinivasan

Heflin

Izatt

et al. 2014

Biomed. Opt. Express

122

View full text Add to dashboard Cite

Accurate quantification of retinal layer thicknesses in mice as seen on optical coherence tomography (OCT) is crucial for the study of numerous ocular and neurological diseases. However, manual segmentation is time-consuming and subjective. Previous attempts to automate this process were limited to high-quality scans from mice with no missing layers or visible pathology. This paper presents an automatic approach for segmenting retinal layers in spectral domain OCT images using sparsity based denoising, support vector machines, graph theory, and dynamic programming (S-GTDP). Results show that this method accurately segments all present retinal layer boundaries, which can range from seven to ten, in wild-type and rhodopsin knockout mice as compared to manual segmentation and has a more accurate performance as compared to the commercial automated Diver segmentation software.

show abstract

Optical Coherence Tomography Machine Learning Classifiers for Glaucoma Detection: A Preliminary Study

Abstract: Automated machine classifiers of OCT data might be useful for enhancing the utility of this technology for detecting glaucomatous abnormality.

Cited by 178 publications

References 34 publications

Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry

Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry

Combining Nerve Fiber Layer Parameters to Optimize Glaucoma Diagnosis with Optical Coherence Tomography

Automatic segmentation of up to ten layer boundaries in SD-OCT images of the mouse retina with and without missing layers due to pathology

Contact Info

Product

Resources

About