Multi-view Learning over Retinal Thickness and Visual Sensitivity on Glaucomatous Eyes

Toshimitsu, Uesaka; Morino, Kazuhide; Sugiura, Hiroki; Kiwaki, Taichi; Murata, Hiroshi; Asaoka, Ryo; Yamanishi, Kenji

doi:10.1145/3097983.3098194

Cited by 8 publications

(21 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Uesaka et al [30] did the first work on estimating VF measured with HFA (SITA-Standard mode 10-2) from RT data. This methodology was based on multi-view learning over two different domains: VF and RT.…”

Section: Related Workmentioning

confidence: 99%

“…Uesaka et al [30] proposed two methods to realize this estimation task based on affine structured non-negative matrix factorization (ASNMF) and a convolutional neural network (CNN). They showed that the ASNMF-based method works better for relatively small data sizes, while the CNN-based method works better for larger data sizes.…”

Section: Introduction 1motivationmentioning

confidence: 99%

“…They also showed that the ASNMF-based method was useful for ophthalmological knowledge discovery. To the best of our knowledge, [30] was the first paper on estimating the VF from RT data.…”

Section: Introduction 1motivationmentioning

confidence: 99%

“…Regarding 1), the root mean squared error (RMSE) for estimation with these methods was 6.8-7.3 in [30], which is insufficient to be used at the real clinical practices. The reason for this inaccuracy is that significant amounts of paired data of VF and RT are required for training, but only imbalanced data are available.…”

Section: Introduction 1motivationmentioning

confidence: 99%

“…Regarding 2), Uesaka et al [30] acquired knowledge about the functional relations between RT and VF using a linear model. The true functional relations, however, can be intrinsically nonlinear [14].…”

Section: Introduction 1motivationmentioning

confidence: 99%

See 4 more Smart Citations

Estimating Glaucomatous Visual Sensitivity from Retinal Thickness with Pattern-Based Regularization and Visualization

Sugiura

Kiwaki

Yousefi

et al. 2018

Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

Conventionally, glaucoma is diagnosed on the basis of visual field sensitivity (VF). However, the VF test is time-consuming, costly, and noisy. Using retinal thickness (RT) for glaucoma diagnosis is currently desirable. Thus, we propose a new methodology for estimating VF from RT in glaucomatous eyes. The key ideas are to use our new methods of pattern-based regularization (PBR) and pattern-based visualization (PBV) with convolutional neural networks (CNNs). PBR effectively conducts supervised learning of RT-VF relations in combination with unsupervised learning from non-paired VF data. We can thereby avoid overfitting of a CNN to small sized data. PBV visualizes functional correspondence between RT and VF with its nonlinearity preserved. We empirically demonstrate with real datasets that a CNN with PBR achieves the highest estimation accuracy to date and that a CNN with PBV is effective for knowledge discovery in an ophthalmological context.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introduction 1motivationmentioning

confidence: 99%

“…They also showed that the ASNMF-based method was useful for ophthalmological knowledge discovery. To the best of our knowledge, [30] was the first paper on estimating the VF from RT data.…”

Section: Introduction 1motivationmentioning

confidence: 99%

Section: Introduction 1motivationmentioning

confidence: 99%

Section: Introduction 1motivationmentioning

confidence: 99%

See 3 more Smart Citations

Estimating Glaucomatous Visual Sensitivity from Retinal Thickness with Pattern-Based Regularization and Visualization

Sugiura

Kiwaki

Yousefi

et al. 2018

Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

AI and Glaucoma

Chen

Wollstein

Schuman

et al. 2021

Artificial Intelligence in Ophthalmology

View full text Add to dashboard Cite

Multimodal Deep Learning Model of Predicting Future Visual Field for Glaucoma Patients

et al. 2023

View full text Add to dashboard Cite

Glaucoma is one of the most common reasons for blindness worldwide, especially in elderly people. Glaucoma can be monitored using visual field (VF) tests. Therefore, predicting the future VF to monitor progression of glaucoma is important. In this paper, we proposed a deep learning model to predict future VF based on previous VF and optical coherence tomography (OCT) images (including thickness map, vertical tomogram, and horizontal tomogram). The image data were analyzed using a ResNet-50 model. Image features and previous VFs were combined, and a long short-term memory (LSTM) network was used to predict future VF. A weighted method was used to detect noisy data. The proposed method was improved when applying weighted loss. The mean absolute error (MAE) was 3.31 ± 1.37, and the root mean square error (RMSE) was 4.58 ± 1.77. The model showed high performance when combining VF data and OCT image data. Furthermore, the model was useful for detecting and re-weighting noisy data.

show abstract

Multi-view Learning over Retinal Thickness and Visual Sensitivity on Glaucomatous Eyes

Cited by 8 publications

References 25 publications

Estimating Glaucomatous Visual Sensitivity from Retinal Thickness with Pattern-Based Regularization and Visualization

Estimating Glaucomatous Visual Sensitivity from Retinal Thickness with Pattern-Based Regularization and Visualization

AI and Glaucoma

Multimodal Deep Learning Model of Predicting Future Visual Field for Glaucoma Patients

Contact Info

Product

Resources

About