![]() ![]() (H) Association between the superior average thickness of the GCA and the superior macular mean sensitivity as measured by HFA. (G) Association between the average thickness of the GCA and the macular mean sensitivity as measured by HFA. (F) Association between the inferior average thickness of the GCA and the inferior macular mean sensitivity as measured by MP-3. (E) Association between the superior average thickness of the GCA and the superior macular mean sensitivity as measured by MP-3. (D) Association between the average thickness of the GCA and the macular mean sensitivity as measured by MP-3. (C) Association between the inferior average thickness of the GCA and the inferior macular mean sensitivity as measured by MAIA TM. (B) Association between the superior average thickness of the GCA and the superior macular mean sensitivity as measured by MAIA TM. (A) Association between the average thickness of the GCA and the macular mean sensitivity as measured by MAIA TM. Among the three VF measurement instruments, however, no significant differences were found for the structure-function relationships.Īll three VF measurement instruments found similar structure-function relationships in the central VF. The highest correlation for the inferior hemiretina (R = 0.687) GCA thickness-VF mean sensitivity was observed by the MP-3. The highest correlation for the global (R = 0.682) and the superior hemiretina (R = 0.594) GCA thickness-VF mean sensitivity was observed by the HFA. The relationship between the GCA thickness and macular sensitivity was examined by Spearman correlation analysis.įor each instrument, statistically significant macular VF sensitivity (dB) and GCA thickness relationships were observed using the decibel scale (R = 0.547-0.687, all P < 0.001). VF and OCT in the retinal view were used to examine both the global relationship between the VF sensitivity and GCA thickness, and the superior hemiretina and inferior hemiretina. All subjects underwent measurements for GCA thickness by Cirrus HD-OCT and static threshold perimetry using MAIATM, MP-3, or HFA. This cross-sectional study examined 73 glaucoma patients and 19 normal subjects. March 10 Available from: 10.1001/ study was conducted in order to compare relationships between the macular visual field (VF) mean sensitivity measured by MAIATM (Macular Integrity Assessment), MP-3, or Humphry field analyzer (HFA) and the ganglion cell and inner plexiform layer (GCA) thicknesses. Course of Glaucomatous Visual Field Loss Across the Entire Perimetric Range. ![]() ![]() Otarola F, Chen A, Morales E, Yu F, Afifi A, Caprioli J. Analysis of progressive change in automated visual fields in glaucoma. Prediction of glaucomatous visual field loss by extrapolation of linear trends. Causes of vision loss worldwide, 1990–2010: a systematic analysis. 10.1136/bjophthalmol-2015-307223īourne RRA, Stevens GA, White RA, Smith JL, Flaxman SR, Price H, et al. Global variations and time trends in the prevalence of primary open angle glaucoma (POAG): a systematic review and meta-analysis. Kapetanakis VV, Chan MPY, Foster PJ, Cook DG, Owen CG, Rudnicka AR. Using unfiltered real-world datasets, deep learning networks show the ability to not only learn spatio-temporal HVF changes but also to generate predictions for future HVFs up to 5.5 years, given only a single HVF. The 100 fully trained models successfully predicted future HVFs in glaucomatous eyes up to 5.5 years in the future with a correlation of 0.92 between the MD of predicted and actual future HVF and an average difference of 0.41 dB. The overall point-wise PMAE for the test set was 2.47 dB (95% CI: 2.45 dB to 2.48 dB), and deep learning showed a statistically significant improvement over linear models. ![]() The best performing model with 20 million trainable parameters, CascadeNet-5, was selected. More than 1.7 million perimetry points were extracted to the hundredth decibel from 32,443 24-2 HVFs. The point-wise mean absolute error (PMAE) and difference in Mean Deviation (MD) between predicted and actual future HVF were calculated. Ten-fold cross validation with a held out test set was used to develop the three main phases of model development: model architecture selection, dataset combination selection, and time-interval model training with transfer learning, to train a deep learning artificial neural network capable of generating a point-wise visual field prediction. To determine if deep learning networks could be trained to forecast future 24-2 Humphrey Visual Fields (HVFs).Īll data points from consecutive 24-2 HVFs from 1998 to 2018 were extracted from a university database. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |