{"title":"Explainable Deep Learning for Glaucomatous Visual Field Prediction: Artifact Correction Enhances Transformer Models.","authors":"Kornchanok Sriwatana, Chanon Puttanawarut, Yanin Suwan, Titipat Achakulvisut","doi":"10.1167/tvst.14.1.22","DOIUrl":"10.1167/tvst.14.1.22","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to develop a deep learning approach that restores artifact-laden optical coherence tomography (OCT) scans and predicts functional loss on the 24-2 Humphrey Visual Field (HVF) test.</p><p><strong>Methods: </strong>This cross-sectional, retrospective study used 1674 visual field (VF)-OCT pairs from 951 eyes for training and 429 pairs from 345 eyes for testing. Peripapillary retinal nerve fiber layer (RNFL) thickness map artifacts were corrected using a generative diffusion model. Three convolutional neural networks and 2 transformer-based models were trained on original and artifact-corrected datasets to estimate 54 sensitivity thresholds of the 24-2 HVF test.</p><p><strong>Results: </strong>Predictive performances were calculated using root mean square error (RMSE) and mean absolute error (MAE), with explainability evaluated through GradCAM, attention maps, and dimensionality reduction techniques. The Distillation with No Labels (DINO) Vision Transformers (ViT) trained on artifact-corrected datasets achieved the highest accuracy (RMSE, 95% confidence interval [CI] = 4.44, 95% CI = 4.07, 4.82 decibel [dB], MAE = 3.46, 95% CI = 3.14, 3.79 dB), and the greatest interpretability, showing improvements of 0.15 dB in global RMSE and MAE (P < 0.05) compared to the performance on original maps. Feature maps and visualization tools indicate that artifacts compromise DINO-ViT's predictive ability but improve with artifact correction.</p><p><strong>Conclusions: </strong>Combining self-supervised ViTs with generative artifact correction enhances the correlation between glaucomatous structures and functions.</p><p><strong>Translational relevance: </strong>Our approach offers a comprehensive tool for glaucoma management, facilitates the exploration of structure-function correlations in research, and underscores the importance of addressing artifacts in the clinical interpretation of OCT.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"22"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11758932/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143024836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mak B Djulbegovic, Henry Bair, David J Taylor Gonzalez, Hiroshi Ishikawa, Gadi Wollstein, Joel S Schuman
{"title":"Artificial Intelligence for Optical Coherence Tomography in Glaucoma.","authors":"Mak B Djulbegovic, Henry Bair, David J Taylor Gonzalez, Hiroshi Ishikawa, Gadi Wollstein, Joel S Schuman","doi":"10.1167/tvst.14.1.27","DOIUrl":"10.1167/tvst.14.1.27","url":null,"abstract":"<p><strong>Purpose: </strong>The integration of artificial intelligence (AI), particularly deep learning (DL), with optical coherence tomography (OCT) offers significant opportunities in the diagnosis and management of glaucoma. This article explores the application of various DL models in enhancing OCT capabilities and addresses the challenges associated with their clinical implementation.</p><p><strong>Methods: </strong>A review of articles utilizing DL models was conducted, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), autoencoders, and large language models (LLMs). Key developments and practical applications of these models in OCT image analysis were emphasized, particularly in the context of enhancing image quality, glaucoma diagnosis, and monitoring progression.</p><p><strong>Results: </strong>CNNs excel in segmenting retinal layers and detecting glaucomatous damage, whereas RNNs are effective in analyzing sequential OCT scans for disease progression. GANs enhance image quality and data augmentation, and autoencoders facilitate advanced feature extraction. LLMs show promise in integrating textual and visual data for comprehensive diagnostic assessments. Despite these advancements, challenges such as data availability, variability, potential biases, and the need for extensive validation persist.</p><p><strong>Conclusions: </strong>DL models are reshaping glaucoma management by enhancing OCT's diagnostic capabilities. However, the successful translation into clinical practice requires addressing major challenges related to data variability, biases, fairness, and model validation to ensure accurate and reliable patient care.</p><p><strong>Translational relevance: </strong>This review bridges the gap between basic research and clinical care by demonstrating how AI, particularly DL models, can markedly enhance OCT's clinical utility in diagnosis, monitoring, and prediction, moving toward more individualized, personalized, and precise treatment strategies.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"27"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11760759/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143034135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Faycal Moujane, Chi Zhang, Robert Knight, John Y Lee, Sophie X Deng, Jie J Zheng
{"title":"Corneal Stromal Stem Cell-Derived Extracellular Vesicles Attenuate ANGPTL7 Expression in the Human Trabecular Meshwork.","authors":"Faycal Moujane, Chi Zhang, Robert Knight, John Y Lee, Sophie X Deng, Jie J Zheng","doi":"10.1167/tvst.14.1.21","DOIUrl":"10.1167/tvst.14.1.21","url":null,"abstract":"<p><strong>Purpose: </strong>Regulating intraocular pressure (IOP), mainly via the trabecular meshwork (TM), is critical in developing glaucoma. Whereas current treatments aim to lower IOP, directly targeting the dysfunctional TM tissue for therapeutic intervention has proven challenging. In our study, we utilized Dexamethasone (Dex)-treated TM cells as a model to investigate how extracellular vesicles (EVs) from immortalized corneal stromal stem cells (imCSSCs) could influence ANGPTL7 and MYOC genes expression within TM cells.</p><p><strong>Methods: </strong>Human TM cell lines were isolated and cultured from donor corneoscleral rims. EVs were purified from imCSSC conditioned media (CM) using size exclusion chromatography and characterized by nanoparticle tracking analysis, transmission electron microscopy (TEM), and ExoView technology. TM cells were treated with either Dex alone or with EVs for 5 days. Quantitative polymerase chain reaction (PCR) was carried out to quantify the mRNA level of MYOC and ANGPTL7.</p><p><strong>Results: </strong>A notable increase in the expression levels of MYOC and ANGPTL7 genes was observed compared with untreated TM cells (control). Furthermore, upon comparing Dex-treated TM cells with those receiving both Dex and EV treatments, a statistically significant reduction in ANGPTL7 expression (P < 0.05) was detected.</p><p><strong>Conclusions: </strong>The present study demonstrates that imCSSCs-derived EVs can effectively decrease the expression of ANGPLT7, a gene associated with fibrosis and implicated in the abnormal elevation of IOP in patients with glaucoma.</p><p><strong>Translational relevance: </strong>Our study shows that imCSSC-derived EVs can specifically target ANGPTL7 expression, making them a promising preclinical therapy for glaucoma.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"21"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11759583/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143060741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Perfusion Capacity as a Predictive Index for Assessing Visual Functional Recovery in Patients With Idiopathic Epiretinal Membrane.","authors":"Jinlian Zhan, Chen Chen, Tong Wang, Qi Zhang, Xia Huang, Lin Lu, Xiujuan Zhao","doi":"10.1167/tvst.14.1.19","DOIUrl":"10.1167/tvst.14.1.19","url":null,"abstract":"<p><strong>Purpose: </strong>This study investigates the association between visual function and retinal vasculature metrics, particularly perfusion capacity (PC), in eyes with idiopathic epiretinal membrane (iERM), using optical coherence tomography angiography (OCTA).</p><p><strong>Methods: </strong>This retrospective study includes 30 eyes from 30 iERM patients who had surgery, with a three-month follow-up period. In addition, 28 eyes from 28 healthy individuals served as a control group. We measured best-corrected visual acuity (BCVA), retinal sensitivity, vascular density (VD), perfusion area (PA), PC, and retinal thickness (RT). OCTA scans assessed both the superficial vascular complex (SVC) and deep vascular complex (DVC) in 3 × 3 mm2 and 6 × 6 mm2 regions. Associations between retinal vasculature metrics and visual outcomes were analyzed.</p><p><strong>Results: </strong>Postoperatively, significant improvements were observed in both BCVA and retinal sensitivity (both P < 0.001). In the SVC layer, VD and PA were significantly decreased (both P < 0.001), whereas PC in the SVC increased significantly (P < 0.001). Higher preoperative PC in the SVC was associated with greater postoperative improvements in retinal sensitivity. Postoperative retinal sensitivity was negatively associated with age (β = -0.53, P = 0.001) and retinal thickness in 3 × 3 mm2 region (β = -0.39, P = 0.013) but positively associated with preoperative retinal sensitivity (β = 1.10, P < 0.001) and PC in the SVC within 3 × 3 mm2 region (β = 0.49, P = 0.023).</p><p><strong>Conclusions: </strong>PC offers a novel approach to evaluating retinal microcirculation and visual prognosis in iERM. Preoperative PC in the SVC serves as a reliable predictive index for predicting postoperative visual recovery.</p><p><strong>Translational relevance: </strong>The PC, as a novel indicator of retinal blood flow, not only reflects the condition of the blood vessels but is also associated with retinal sensitivity.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"19"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11756611/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143012058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Muhammad Usman Jamil, Jungeun Won, Stefan B Ploner, Anna Marmalidou, Hiroyuki Takahashi, Stephanie Kaiser, Yunchan Hwang, Omar Abu-Qamar, Antonio Yaghy, Andre J Witkin, Peter Y Zhao, Shilpa Desai, Jay S Duker, Andreas Maier, James G Fujimoto, Nadia K Waheed
{"title":"High-Resolution OCT Reveals Age-Associated Variation in the Region Posterior to the External Limiting Membrane.","authors":"Muhammad Usman Jamil, Jungeun Won, Stefan B Ploner, Anna Marmalidou, Hiroyuki Takahashi, Stephanie Kaiser, Yunchan Hwang, Omar Abu-Qamar, Antonio Yaghy, Andre J Witkin, Peter Y Zhao, Shilpa Desai, Jay S Duker, Andreas Maier, James G Fujimoto, Nadia K Waheed","doi":"10.1167/tvst.14.1.16","DOIUrl":"10.1167/tvst.14.1.16","url":null,"abstract":"<p><strong>Purpose: </strong>To evaluate visibility of a sub-band posterior to the external limiting membrane (ELM) and assess its age-associated variation.</p><p><strong>Methods: </strong>In a retrospective cross-sectional study, normal eyes were imaged using a high-resolution spectral-domain optical coherence tomography (SD-OCT) prototype (2.7-µm axial resolution). Volume fusion of six sequential scans (each 500 × 500 A-scans over 6 mm × 6 mm) was performed in the motion correction and volume reconstruction in OCT (MoReOCT) framework to enhance feature visibility in OCT. The subjects were divided into three groups: young (21-40 years old), middle (41-60 years old), and older (>60 years old). Three expert graders assessed the visibility of the sub-band on B-scans, and its A-scan intensity relative to ELM intensity (peak intensity ratio) was measured.</p><p><strong>Results: </strong>Forty-four eyes of 44 subjects were imaged. The sub-band, tentatively attributed to the photoreceptor myoid, can be visualized under high-resolution OCT. The B-scan gradings showed that sub-band visibility increased with age (visible in 16.7%, 47.2%, and 66.7% of the young, middle, and older age groups, respectively). The gradings were statistically different among age groups at 1 mm and 2 mm nasal and 1 mm and 2 mm temporal (P < 0.04) from the foveal center. Similarly, the mean peak intensity ratios of the sub-band to the ELM were 71.6%, 77.5%, and 85.2% in the young, middle, and older age groups, respectively, and were positively correlated with age at 1 mm temporal (P = 0.012) and 2 mm temporal (P < 0.001).</p><p><strong>Conclusions: </strong>High-resolution OCT, combined with advanced volume fusion, enables visualization of the photoreceptor myoid and investigation of its age-associated variations.</p><p><strong>Translational relevance: </strong>Investigating the sub-band can advance our understanding of photoreceptors and their association with aging and disease pathogenesis.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"16"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11741063/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142984974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Annabelle Gallois-Bernos, Lichun Zhong, Mingqi Bai, Caroline A Blackie
{"title":"A Novel Rabbit Model of Meibomian Gland Dysfunction-Induced Dry Eye.","authors":"Annabelle Gallois-Bernos, Lichun Zhong, Mingqi Bai, Caroline A Blackie","doi":"10.1167/tvst.14.1.28","DOIUrl":"10.1167/tvst.14.1.28","url":null,"abstract":"<p><strong>Purpose: </strong>The objective of this three-phase study was to develop a model of mild to moderate evaporative dry eye to be used to evaluate tear film stability endpoints during product development.</p><p><strong>Methods: </strong>Rabbits were sedated prior to ophthalmic cautery of meibomian gland orifices. The orifices of eyelid meibomian glands were half-cauterized (to yield obstruction of every other meibomian gland orifices), fully cauterized (to yield obstruction of all meibomian gland orifices), or untreated. The primary outcome measures were model repeatability, tear film break-up time (TBUT), non-invasive tear break-up time (NIBUT), and model confirmation with daily artificial tears. Other examinations included Draize score, tear production, fluorescein staining, eschar, and histopathology.</p><p><strong>Results: </strong>Cauterization was well tolerated in all phases. In phase I, TBUT was significantly decreased compared to baseline on days 14, 21, 28, and 35 in fully cauterized meibomian gland orifices but not in half-cauterized meibomian gland orifices. In phase II, both NIBUT and TBUT were similarly and significantly reduced in the fully cauterized meibomian gland orifices compared to the control eyes on days 7, 9, 14, and 28. In the confirmation phase, the administration of eye drops significantly improved NIBUT over the course of the study.</p><p><strong>Conclusions: </strong>A rabbit model of dry eye was successfully and safely created through the obstruction of meibomian gland orifices by cautery, which yielded a significant reduction in tear film stability. A quantifiable benefit of artificial tears relative to untreated control was demonstrated within the model.</p><p><strong>Translational relevance: </strong>When the model is deployed, the utility and efficacy of therapeutic formulations can be evaluated.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"28"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11781322/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143060736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Immanuel P Seitz, Fabian Wozar, Guy Alex Ochakovski, Felix F Reichel, Sven Korte, Birgit Korbmacher, Barbara Wilhelm, Daniela Süsskind, Karl-Ulrich Bartz-Schmidt, M Dominik Fischer, Tobias Peters
{"title":"Ocular Safety and Toxicology of Subretinal Gene Therapy With rAAV.hPDE6A in Nonhuman Primates.","authors":"Immanuel P Seitz, Fabian Wozar, Guy Alex Ochakovski, Felix F Reichel, Sven Korte, Birgit Korbmacher, Barbara Wilhelm, Daniela Süsskind, Karl-Ulrich Bartz-Schmidt, M Dominik Fischer, Tobias Peters","doi":"10.1167/tvst.14.1.29","DOIUrl":"10.1167/tvst.14.1.29","url":null,"abstract":"<p><strong>Purpose: </strong>Reports of gene therapy-associated retinal atrophies and inflammation have highlighted the importance of preclinical safety assessments of adeno-associated virus (AAV) vector systems. We evaluated in nonhuman primates (NHPs) the ocular safety and toxicology of a novel AAV gene therapy targeting retinitis pigmentosa caused by mutations in PDE6A, which has since been used in a phase I/II clinical trial (NCT04611503).</p><p><strong>Methods: </strong>A total of 34 healthy cynomolgus animals (Macaca fascicularis) were treated with subretinal injections of rAAV.hPDE6A and followed over 13 weeks. Three dose levels (low: 1 × 1011, intermediate: 5 × 1011, and high: 1 × 1012 vector genomes [vg]) were compared to sham-injected controls. Safety and toxicity were determined using ophthalmic examinations, electroretinography, ocular histology, and retinal imaging.</p><p><strong>Results: </strong>At the low and intermediate doses, inflammation was mild, electroretinography response was unimpeded, and histology results were in line with surgically induced changes. In contrast, three high-dose animals displayed atrophic changes of the retina and abnormalities in electroretinography, which were considered test article related and adverse.</p><p><strong>Conclusions: </strong>A single subretinal injection of up to 5 × 1011 vg was well tolerated, and a 10-fold lower dose of 5 × 1010 vg was chosen as the starting dose for the ongoing phase I/II clinical trial. Atrophic retinal changes and abnormalities in electroretinography emerged as dose-limiting findings in the high-dose cohort.</p><p><strong>Translational relevance: </strong>This study demonstrates that treatment candidate rAAV.PDE6A was well tolerated in NHPs. Occurrence of retinal atrophy as a dose-limiting finding highlights the importance of further study into the mechanisms of atrophy induction after retinal gene therapy.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"29"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11781330/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143060749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ines D Nagel, Anna Heinke, Akshay P Agnihotri, Shaden Yassin, Lingyun Cheng, Andrew S Camp, Nathan L Scott, Fritz Gerald P Kalaw, Shyamanga Borooah, Dirk-Uwe G Bartsch, Arthur J Mueller, Nehal Mehta, William R Freeman
{"title":"Comparison of a Novel Ultra-Widefield Three-Color Scanning Laser Ophthalmoscope to Other Retinal Imaging Modalities in Chorioretinal Lesion Imaging.","authors":"Ines D Nagel, Anna Heinke, Akshay P Agnihotri, Shaden Yassin, Lingyun Cheng, Andrew S Camp, Nathan L Scott, Fritz Gerald P Kalaw, Shyamanga Borooah, Dirk-Uwe G Bartsch, Arthur J Mueller, Nehal Mehta, William R Freeman","doi":"10.1167/tvst.14.1.11","DOIUrl":"10.1167/tvst.14.1.11","url":null,"abstract":"<p><strong>Purpose: </strong>To compare the assessment of clinically relevant retinal and choroidal lesions as well as optic nerve pathologies using a novel three-wavelength ultra-widefield (UWF) scanning laser ophthalmoscope with established retinal imaging techniques for ophthalmoscopic imaging.</p><p><strong>Methods: </strong>Eighty eyes with a variety of retinal and choroidal lesions were assessed on the same time point using Topcon color fundus photography (CFP) montage, Optos red/green (RG), Heidelberg SPECTRALIS MultiColor 55-color montage (MCI), and novel Optos red/green/blue (RGB). Paired images of the optic nerve, retinal, or choroidal lesions were initially diagnosed based on CFP imaging. The accuracy of the imaging was then evaluated in comparison to CFP using a grading scale ranging from -1 (losing imaging information) to +1 (gaining imaging information).</p><p><strong>Results: </strong>Eighty eyes of 43 patients with 116 retinal or choroidal pathologies, as well as 59 eyes with optic nerve imaging using CFP, MCI, RG, and RGB, were included in this study. Across all subgroups, RGB provided significantly more accurate clinical imaging with CFP as ground truth and compared to other modalities. This was true comparing RGB to both RG (P = 0.0225) and MCI (P < 0.001) overall. Although RGB provided more accurate clinical information overall, it was inferior to RG for melanocytic choroidal lesions (P = 0.011).</p><p><strong>Conclusions: </strong>RGB can be considered as a useful tool to detect characteristics of central, midperipheral, and peripheral retinal lesions. Regarding melanocytic choroidal lesions, RGB was inferior to RG, and MCI was inferior to both RG and RGB modalities due to color changes.</p><p><strong>Translational relevance: </strong>Traditional retinal ultra-widefield imaging uses two wavelengths. Here, we evaluated three wavelengths for ultra-widefield imaging. We examined new optics (basic science) effect on patient imaging (clinical care).</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"11"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11737455/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142972304","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"In Vitro Analysis of Pressure Resistance in the Paul Glaucoma Implant and Ahmed ClearPath 250 With and Without Polypropylene Thread Inside the Tube.","authors":"Andi Masdipa, Sachiko Kaidzu, Masaki Tanito","doi":"10.1167/tvst.14.1.2","DOIUrl":"10.1167/tvst.14.1.2","url":null,"abstract":"<p><strong>Purpose: </strong>Pressure resistance characteristics of the Paul glaucoma implant (PGI) and Ahmed ClearPath 250 (ACP), with and without the insertion of polypropylene thread in their tubes, were evaluated.</p><p><strong>Methods: </strong>The in vitro flow pressure was evaluated at varying flow rates, both with and without threads (6-0 for PGI and 4-0 or 3-0 for ACP). Cross-sectional areas of the tube lumen and thread were measured to calculate pressure resistance using the Hagen-Poiseuille equation.</p><p><strong>Results: </strong>For the PGI without the thread, the pressure remained relatively low and constant across all flow rates. In contrast, with the insertion of a 6-0 thread, there was a significant increase in pressure resistance, with the pressure increasing from 7.5 mm Hg at 1 µL/min to 43.8 mm Hg at 5 µL/min. For the ACP, the pressure resistance remained relatively constant across all flow rates without a thread and with either 4-0 or 3-0 threads. However, the pressure was higher with 3-0 threads compared with a 4-0 thread. The actual measured pressures agreed well with theoretical values in the no-thread conditions, but were consistently higher than the theoretical values in the threaded conditions.</p><p><strong>Conclusions: </strong>Inserting polypropylene threads into the tubes of nonvalved glaucoma drainage devices significantly affects pressure resistance with various degree.</p><p><strong>Translational relevance: </strong>PGI with a 6-0 polypropylene thread may not require external tube ligation to prevent hypotony, whereas ACP with a 4-0 thread likely requires additional ligation. Using a 3-0 thread in ACP may enhance pressure resistance sufficiently to avoid tube ligation, but this requires careful clinical consideration.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"2"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11717134/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142955736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rehana Khan, Vinod Maseedupally, Kaveri A Thakoor, Rajiv Raman, Maitreyee Roy
{"title":"Noninvasive Anemia Detection and Hemoglobin Estimation from Retinal Images Using Deep Learning: A Scalable Solution for Resource-Limited Settings.","authors":"Rehana Khan, Vinod Maseedupally, Kaveri A Thakoor, Rajiv Raman, Maitreyee Roy","doi":"10.1167/tvst.14.1.20","DOIUrl":"10.1167/tvst.14.1.20","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to develop and validate a deep-learning model for noninvasive anemia detection, hemoglobin (Hb) level estimation, and identification of anemia-related retinal features using fundus images.</p><p><strong>Methods: </strong>The dataset included 2265 participants aged 40 years and above from a population-based study in South India. The dataset included ocular and systemic clinical parameters, dilated retinal fundus images, and hematological data such as complete blood counts and Hb concentration levels. Eighty percent of the dataset was used for algorithm development and 20% for validation. A deep-convolutional neural network, utilizing VGG16, ResNet50, and InceptionV3 architectures, was trained to predict anemia and estimate Hb levels. Sensitivity, specificity, and accuracy were calculated, and receiver operating characteristic (ROC) curves were generated for comparison with clinical anemia data. GradCAM saliency maps highlighted regions linked to anemia and image processing techniques to quantify anemia-related features.</p><p><strong>Results: </strong>For predicting anemia, the InceptionV3 model demonstrated the best performance, achieving 98% accuracy, 99% sensitivity, 97% specificity, and an area under the curve (AUC) of 0.98 (95% confidence interval [CI] = 0.97-0.99). For estimating Hb levels, the mean absolute error for the InceptionV3 model was 0.58 g/dL (95% CI = 0.57-0.59 g/dL). The model focused on the area around the optic disc and the neighboring retinal vessels, revealing that anemic subjects exhibited significantly increased vessel tortuosity and reduced vessel density (P < 0.001), with variable effects on vessel thickness.</p><p><strong>Conclusions: </strong>The InceptionV3 model accurately predicted anemia and Hb levels, highlighting the potential of deep learning and vessel analysis for noninvasive anemia detection.</p><p><strong>Translational relevance: </strong>The proposed method offers the possibility to quantitatively predict hematological parameters in a noninvasive manner.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 1","pages":"20"},"PeriodicalIF":2.6,"publicationDate":"2025-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11771037/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143024875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}