Search

Vision Transformers have recently emerged as a competitive architecture in image classification. The tremendous popularity of this model and its variants comes from its high performance and its ability to produce interpretable predictions. However, both of these characteristics remain to be assessed in depth on retinal images. This study proposes a thorough performance evaluation of several Transformers compared to traditional Convolutional Neural Network (CNN) models for retinal disease classification. Special attention is given to multi-modality imaging (fundus and OCT) and generalization to external data. In addition, we propose a novel mechanism to generate interpretable predictions via attribution maps. Existing attribution methods from Transformer models have the disadvantage of producing low-resolution heatmaps. Our contribution, called Focused Attention, uses iterative conditional patch resampling to tackle this issue. By means of a survey involving four retinal specialists, we validated both the superior interpretability of Vision Transformers compared to the attribution maps produced from CNNs and the relevance of Focused Attention as a lesion detector., Competing Interests: Declaration of Competing Interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Clément Playout reports financial support was provided by Diabetes Action Canada., (Copyright © 2022 Elsevier B.V. All rights reserved.)

Ocular rigidity (OR) is thought to play a role in the pathogenesis of glaucoma, but the lack of reliable non-invasive measurements has been a major technical challenge. We recently developed a clinical method using optical coherence tomography time-lapse imaging and automated choroidal segmentation to measure the pulsatile choroidal volume change (ΔV) and calculate OR using Friedenwald's equation. Here we assess the validity and repeatability of this non-invasive technique. We also propose an improved mathematical model of choroidal thickness to extrapolate ΔV from the pulsatile submacular choroidal thickness change more accurately. The new mathematical model uses anatomical data accounting for the choroid thickness near the equator. The validity of the technique was tested by comparing OR coefficients obtained using our non-invasive method (OR OCT ) and those obtained with an invasive procedure involving intravitreal injections of Bevacizumab (OR IVI ) in 12 eyes. Intrasession and intersession repeatability was assessed for 72 and 8 eyes respectively with two consecutive measurements of OR. Using the new mathematical model, we obtained OR values which are closer to those obtained using the invasive procedure and previously reported techniques. A regression line was calculated to predict the OR IVI based on OR OCT , such that OR IVI  = 0.655 × OR OCT . A strong correlation between OR OCT and OR IVI was found, with a Spearman coefficient of 0.853 (p < 0.001). The intraclass correlation coefficient for intrasession and intersession repeatability was 0.925, 95% CI [0.881, 0.953] and 0.950, 95% CI [0.763, 0.990] respectively. This confirms the validity and good repeatability of OR measurements using our non-invasive clinical method., (Copyright © 2019 Elsevier Ltd. All rights reserved.)

Obtaining the complete segmentation map of retinal lesions is the first step toward an automated diagnosis tool for retinopathy that is interpretable in its decision-making. However, the limited availability of ground truth lesion detection maps at a pixel level restricts the ability of deep segmentation neural networks to generalize over large databases. In this paper, we propose a novel approach for training a convolutional multi-task architecture with supervised learning and reinforcing it with weakly supervised learning. The architecture is simultaneously trained for three tasks: segmentation of red lesions and of bright lesions, those two tasks done concurrently with lesion detection. In addition, we propose and discuss the advantages of a new preprocessing method that guarantees the color consistency between the raw image and its enhanced version. Our complete system produces segmentations of both red and bright lesions. The method is validated at the pixel level and per-image using four databases and a cross-validation strategy. When evaluated on the task of screening for the presence or absence of lesions on the Messidor image set, the proposed method achieves an area under the ROC curve of 0.839, comparable with the state-of-the-art.

Purpose: To evaluate the rate of sclerotomy suturing in a large series of primary 23-gauge vitrectomies., Methods: Retrospective chart review of 23-gauge transconjunctival pars plana vitrectomies performed between 2008 and 2012 in a multisurgeon retina practice. Patients with a history of retinal surgery, combination of vitrectomy and scleral buckling surgery, or intraoperative conversion of a sclerotomy to 20-gauge were excluded from the study. Vitrectomies were divided into two types according to their degree of complexity and intraoperative use of instrumentation. Demographic and clinical data were gathered from the chart. Location and number of sutured sclerotomies were obtained from the operative report., Results: A total of 589 eyes from 569 patients were included in this study. At least 1 sclerotomy was sutured in 227 eyes (38.5%): 69 (30.4%) had 1 sutured sclerotomy, 52 (22.9%) had 2 sutured sclerotomies, and 106 (46.7%) had 3 sutured sclerotomies. Sclerotomy suturing was significantly related to the surgical technique (P < 0.001), type of tamponade agent (P = 0.048), and complexity of the vitrectomy (P < 0.001). The incidence of postoperative hypotony at Day 1 was 2.2% and all cases returned to normal intraocular pressure at 1 week. There were no recognized complications related to sclerotomies regardless of suturing., Conclusion: The factors that increased the rate of sclerotomy suturing were the absence of tamponade agent and higher complexity of the procedure. There was a high variability in the rate of suturing between the surgeons.

Background: Code-free deep learning (CFDL) is a novel tool in artificial intelligence (AI). This study directly compared the discriminative performance of CFDL models designed by ophthalmologists without coding experience against bespoke models designed by AI experts in detecting retinal pathologies from optical coherence tomography (OCT) videos and fovea-centered images. Methods: Using the same internal dataset of 1,173 OCT macular videos and fovea-centered images, model development was performed simultaneously but independently by an ophthalmology resident (CFDL models) and a postdoctoral researcher with expertise in AI (bespoke models). We designed a multi-class model to categorize video and fovea-centered images into five labels: normal retina, macular hole, epiretinal membrane, wet age-related macular degeneration and diabetic macular edema. We qualitatively compared point estimates of the performance metrics of the CFDL and bespoke models. Results: For videos, the CFDL model demonstrated excellent discriminative performance, even outperforming the bespoke models for some metrics: area under the precision-recall curve was 0.984 (vs. 0.901), precision and sensitivity were both 94.1% (vs. 94.2%) and accuracy was 94.1% (vs. 96.7%). The fovea-centered CFDL model overall performed better than video-based model and was as accurate as the best bespoke model. Conclusion: This comparative study demonstrated that code-free models created by clinicians without coding expertise perform as accurately as expert-designed bespoke models at classifying various retinal pathologies from OCT videos and images. CFDL represents a step forward towards the democratization of AI in medicine, although its numerous limitations must be carefully addressed to ensure its effective application in healthcare. [ABSTRACT FROM AUTHOR]