Selective fusion of structural and functional data for improved glaucoma detection

This work proposes novel selective feature fusion of structural and functional data for improved glaucoma detection. The structural data such as retinal nerve fiber layer (RNFL) thickness measurement acquired by scanning laser polarimetry (SLP) is fused with the functional visual field (VF) measurement recorded from the standard automated perimetry (SAP) test. The proposed selective feature fusion exploits correspondence between structural and functional data obtained over multiple sectors. The correlation coefficients for corresponding structural-function sector pairs are used as weights in subsequent feature selection. The sectors are ranked according to the correlation coefficients and the first four highly ranked sectors are retained. Following our prior work, fractal analysis (FA) features for both structural and functional data are obtained and fused for each selected sectors respectively. These fused FA features are then used for glaucoma detection. The novelty of this work stems from (i) locating structure-functional sectoral correspondence; (ii) selecting only a few interesting sector pairs using correlation coefficient between structure-function data; (iii) obtaining novel FA features from these pairs; and (iv) fusing these features for glaucoma detection. Such a method is distinctively different from other existing methods that exploit structure-function models in that structure-function sectoral correspondences have been weighted and, based on such weights, only portions of the sectors are retained for subsequent fusion and classification of structural and functional features. For statistical analysis of the glaucoma detection results, sensitivity, specificity and area under receiver operating characteristic curve (AUROC) are calculated. Performance comparison is obtained with those of existing feature-based techniques such as wavelet-Fourier analysis (WFA) and fast-Fourier analysis (FFA). Comparisons of AUROC values show that our novel selective feature fusion method for discrimination of glaucomatous and ocular normal patients slightly outperforms other existing techniques with AUROCs of 0.98, 0.98 and 0.99 for WFA, FFA and FA respectively.

Download Full-text

Elucidating Macular Structure-Function Correlations in Glaucoma

10.21203/rs.3.rs-1226007/v1 ◽

2022 ◽

Author(s):

Sara Giammaria ◽

Glen Sharpe ◽

Dyachojk Oksana ◽

Paul Rafuse ◽

Shuba Lesya ◽

...

Keyword(s):

Structure Function ◽

Functional Data ◽

Structural Damage ◽

Structural Data ◽

Automated Perimetry ◽

Standard Automated Perimetry ◽

Measurement Variability ◽

Maximum Value ◽

Precise Mapping ◽

Generate Vector

Abstract Correlation between structural data from optical coherence tomography (OCT) and functional data from the visual field (VF) may be suboptimal because of poor mapping of OCT measurement locations to VF test stimuli. We tested the hypothesis that stronger structure-function correlations in the macula can be achieved with fundus-tracking perimetery, by precisely mapping OCT measurements to VF sensitivity at the same location. The conventional 64 superpixel (3°x3°) OCT grid was mapped to VF sensitivities averaged in 40 corresponding VF units with standard automated perimetry (conventional mapped approach, CMA) in 38 glaucoma patients and 10 healthy subjects. Similarly, a 144 superpixel (2°x2°) OCT grid was mapped to each of the 68 VF locations with fundus-tracking perimetry (localized mapped approach, LMA). For each approach, the correlation between sensitivity at each VF unit and OCT superpixel was computed and the maximum value used to generate vector maps. CMA yielded significantly higher structure-function correlations compared to LMA. Only 20% of the vectors with CMA and <5% with LMA were within corresponding mapped OCT superpixels, while most were directed towards loci with structural damage. Measurement variability and patterns of glaucomatous damage are more likely to affect the correlations rather than precise mapping of VF stimuli.

Download Full-text

Investigating the structure‐function relationship using Goldmann V standard automated perimetry where glaucomatous damage is advanced

Ophthalmic and Physiological Optics ◽

10.1111/opo.12643 ◽

2019 ◽

Vol 39 (6) ◽

pp. 441-450

Author(s):

Mieko Yanagisawa ◽

Hiroshi Murata ◽

Masato Matsuura ◽

Yuri Fujino ◽

Kazunori Hirasawa ◽

...

Keyword(s):

Structure Function ◽

Automated Perimetry ◽

Standard Automated Perimetry ◽

Glaucomatous Damage ◽

Structure Function Relationship ◽

Function Relationship

Download Full-text

Structure-Function Modeling of Optical Coherence Tomography and Standard Automated Perimetry in the Retina of Patients with Autosomal Dominant Retinitis Pigmentosa

PLoS ONE ◽

10.1371/journal.pone.0148022 ◽

2016 ◽

Vol 11 (2) ◽

pp. e0148022 ◽

Cited By ~ 24

Author(s):

Travis B. Smith ◽

Maria Parker ◽

Peter N. Steinkamp ◽

Richard G. Weleber ◽

Ning Smith ◽

...

Keyword(s):

Optical Coherence Tomography ◽

Retinitis Pigmentosa ◽

Structure Function ◽

Autosomal Dominant ◽

Automated Perimetry ◽

Optical Coherence ◽

Standard Automated Perimetry ◽

Autosomal Dominant Retinitis Pigmentosa

Download Full-text

Video Summarization Based on Multimodal Features

International Journal of Multimedia Data Engineering and Management ◽

10.4018/ijmdem.2020100104 ◽

2020 ◽

Vol 11 (4) ◽

pp. 60-76

Author(s):

Yu Zhang ◽

Ju Liu ◽

Xiaoxi Liu ◽

Xuesong Gao

Keyword(s):

Feature Fusion ◽

Video Summarization ◽

Video Clip ◽

Rank Correlation ◽

Correlation Coefficients ◽

Ground Truth ◽

Video Summary ◽

Video Feature ◽

Multimodal Features ◽

Feature Sequence

In this manuscript, the authors present a keyshots-based supervised video summarization method, where feature fusion and LSTM networks are used for summarization. The framework can be divided into three folds: 1) The authors formulate video summarization as a sequence to sequence problem, which should predict the importance score of video content based on video feature sequence. 2) By simultaneously considering visual features and textual features, the authors present the deep fusion multimodal features and summarize videos based on recurrent encoder-decoder architecture with bi-directional LSTM. 3) Most importantly, in order to train the supervised video summarization framework, the authors adopt the number of users who decided to select current video clip in their final video summary as the importance scores and ground truth. Comparisons are performed with the state-of-the-art methods and different variants of FLSum and T-FLSum. The results of F-score and rank correlation coefficients on TVSum and SumMe shows the outstanding performance of the method proposed in this manuscript.

Download Full-text

Comparing the Structure–Function Relationship at the Macula With Standard Automated Perimetry and Microperimetry

Investigative Opthalmology & Visual Science ◽

10.1167/iovs.15-17922 ◽

2015 ◽

Vol 56 (13) ◽

pp. 8063 ◽

Cited By ~ 16

Author(s):

Harsha L. Rao ◽

Manideepak Januwada ◽

Raza S. M. Hussain ◽

Lalitha N. Pillutla ◽

Viquar U. Begum ◽

...

Keyword(s):

Structure Function ◽

Automated Perimetry ◽

Standard Automated Perimetry ◽

Structure Function Relationship ◽

Function Relationship

Download Full-text

Mapping the Structure-Function Relationship in Glaucoma and Healthy Patients Measured with Spectralis OCT and Humphrey Perimetry

Journal of Ophthalmology ◽

10.1155/2018/1345409 ◽

2018 ◽

Vol 2018 ◽

pp. 1-12

Author(s):

Laia Jaumandreu ◽

Francisco J. Muñoz–Negrete ◽

Noelia Oblanca ◽

Gema Rebolleda

Keyword(s):

Structure Function ◽

Visual Fields ◽

Macular Thickness ◽

Automated Perimetry ◽

Humphrey Perimetry ◽

Glaucoma Diagnosis ◽

Structure Function Relationship ◽

Statistical Approaches ◽

Function Relationship ◽

Spectralis Oct

Purpose. To study the structure-function relationship in glaucoma and healthy patients assessed with Spectralis OCT and Humphrey perimetry using new statistical approaches. Materials and Methods. Eighty-five eyes were prospectively selected and divided into 2 groups: glaucoma (44) and healthy patients (41). Three different statistical approaches were carried out: (1) factor analysis of the threshold sensitivities (dB) (automated perimetry) and the macular thickness (μm) (Spectralis OCT), subsequently applying Pearson’s correlation to the obtained regions, (2) nonparametric regression analysis relating the values in each pair of regions that showed significant correlation, and (3) nonparametric spatial regressions using three models designed for the purpose of this study. Results. In the glaucoma group, a map that relates structural and functional damage was drawn. The strongest correlation with visual fields was observed in the peripheral nasal region of both superior and inferior hemigrids (r=0.602 and r=0.458, resp.). The estimated functions obtained with the nonparametric regressions provided the mean sensitivity that corresponds to each given macular thickness. These functions allowed for accurate characterization of the structure-function relationship. Conclusions. Both maps and point-to-point functions obtained linking structure and function damage contribute to a better understanding of this relationship and may help in the future to improve glaucoma diagnosis.

Download Full-text

Structure–Function Correlations in Glaucoma Using Matrix and Standard Automated Perimetry Versus Time-Domain and Spectral-Domain OCT Devices

Investigative Opthalmology & Visual Science ◽

10.1167/iovs.13-13664 ◽

2014 ◽

Vol 55 (5) ◽

pp. 3074 ◽

Cited By ~ 13

Author(s):

Luciano Moreira Pinto ◽

Elaine Fiod Costa ◽

Luiz Alberto S. Melo ◽

Paula Blasco Gross ◽

Eduardo Toshio Sato ◽

...

Keyword(s):

Structure Function ◽

Time Domain ◽

Spectral Domain ◽

Automated Perimetry ◽

Standard Automated Perimetry ◽

Spectral Domain Oct

Download Full-text

Combined structure–function analysis in glaucoma screening

British Journal of Ophthalmology ◽

10.1136/bjophthalmol-2021-319178 ◽

2021 ◽

pp. bjophthalmol-2021-319178

Author(s):

Elina Karvonen ◽

Katri Stoor ◽

Marja Luodonpää ◽

Pasi Hägg ◽

Ilmari Leiviskä ◽

...

Keyword(s):

Structure Function ◽

Function Analysis ◽

Classification Tree ◽

Visual Fields ◽

Automated Perimetry ◽

Middle Aged ◽

Predictive Values ◽

Aged Population ◽

Glaucoma Screening ◽

Sd Oct

AimTo assess the applicability of a structure–function (S-F) analysis combining spectral-domain optical coherence tomography (SD-OCT) and standard automated perimetry (SAP) in glaucoma screening in a middle-aged population.MethodsA randomised sample of 3001 Caucasian participants aged 45–49 years of the Northern Finland Birth Cohort Eye Study was examined. We performed an eye examination, including 24–2 SAP, optic nerve head (ONH) and retinal nerve fibre layer (RNFL) photography and SD-OCT of the peripapillary RNFL. The S-F report was generated by Forum Glaucoma Workplace software. OCT, SAP and the S-F analysis were evaluated against clinical glaucoma diagnosis, that is, the positive ‘2 out of 3’ rule based on the clinician’s evaluation of ONH and RNFL photographs and visual fields (VFs).ResultsAt a specificity of 97.5%, the sensitivity for glaucomatous damage was 26% for abnormal OCT, 35% for SAP and 44% for S-F analysis. Estimated areas under the curve were 0.74, 0.85 and 0.76, and the corresponding positive predictive values were 8 %, 10% and 12%, respectively. By applying a classification tree approach combining OCT, SAP and defect localisation data, a sensitivity of 77% was achieved at 90% specificity. In a localisation analysis of glaucomatous structural and functional defects, the correlation with glaucoma increased significantly if the abnormal VF test points were located on borderline or abnormal OCT zones.ConclusionSAP performs slightly better than OCT in glaucoma screening of middle-aged population. However, the diagnostic capability can be improved by S-F analysis.

Download Full-text

Applying a New Automated Perimetry Pattern Based on the Stimulus Distribution of the Multifocal ERG to Improve Structure-Function Investigation in Glaucoma

Journal of Ophthalmology ◽

10.1155/2017/8780934 ◽

2017 ◽

Vol 2017 ◽

pp. 1-11 ◽

Cited By ~ 1

Author(s):

Lívia M. Brandão ◽

Matthias Monhart ◽

Andreas Schötzau ◽

Anna A. Ledolter ◽

Anja M. Palmowski-Wolfe

Keyword(s):

Optical Coherence Tomography ◽

Structure Function ◽

Automated Perimetry ◽

Optical Coherence ◽

Fiber Layer ◽

Rnfl Thickness ◽

Significant Difference ◽

Retinal Structure ◽

One Year

Purpose. To validate a new automated perimetry pattern (mf103 pattern) for the investigation of retinal structure-function relationships in glaucoma in comparison to the standard G2 pattern and to relate either field’s performance to optical coherence tomography (OCT). Methods. Automated perimetry data from the mfERG103 pattern were compared with the standard G2 pattern in glaucoma patients (18) and controls (15). The results of both (mean defect (MD) and mean sensitivity (MS)) were compared with optical coherence tomography (OCT): retinal nerve fiber layer (RNFL) thickness, macular thickness (mT), and ganglion cell analysis (GCIPL). Nine patients were followed up after one year. Results. G2 pattern and mf103 pattern did not differ significantly in MD or MS. The mf103 pattern associated significantly with more RNFL sectors in both MD and MS (p<0.01 and p<0.05, resp.). GCIPL thickness was not significantly associated with either SAP protocols. Both protocols remained comparable after one-year follow-up. Conclusions. G2 and mf103 pattern can both differentiate patients from controls with no significant difference in performance. RNFL thickness defects correlated better with mf103 than G2 with POAG. The mfERG-103 perimetry pattern can be used to establish structure-function correlations in glaucoma and may enable a more direct comparison with objective electrophysiological data.

Download Full-text