To quantitatively measure hyperreflective foci (HRF) during the progression of geographic atrophy (GA) secondary to age-related macular degeneration (AMD) using deep learning (DL) and investigate association with local global growth GA.Eyes GA were prospectively included. Spectral-domain optical coherence tomography (SDOCT) fundus autofluorescence images acquired every 6 months. A 500-μm-wide junctional zone adjacent border was delineated HRF quantified a validated DL algorithm....

To investigate the therapeutic effect of intravitreal pegcetacoplan on inhibition photoreceptor (PR) loss and thinning in geographic atrophy (GA) conventional spectral-domain OCT (SD-OCT) imaging by deep learning-based automated PR quantification.Post hoc analysis a prospective, multicenter, randomized, sham (SM)-controlled, masked phase II trial investigating safety efficacy for treatment GA because age-related macular degeneration.Study eyes 246 patients, randomized 1:1:1 to monthly (AM),...

ObjectiveTo identify the individual progression of geographic atrophy (GA) lesions from baseline optical coherence tomography (OCT) images in patients clinical routine.DesignClinical evaluation a deep-learning based algorithmSubjects184 eyes 100 consecutively enrolled patientsMethodsOCT and fundus autofluorescence (FAF) (both Spectralis, Heidelberg Engineering, Heidelberg, Germany) with GA secondary to age-related macular degeneration routine care were used for model validation. FAF...

In this paper, we introduce a Bayesian deep learning based model for segmenting the photoreceptor layer in pathological OCT scans. Our architecture provides accurate segmentations of and produces pixel-wise epistemic uncertainty maps that highlight potential areas pathologies or segmentation errors. We empirically evaluated approach two sets scans patients with age-related macular degeneration, retinal vein oclussion diabetic edema, improving performance baseline U-Net both terms Dice index...

To investigate quantitative differences in fluid volumes between subretinal (SRF)-tolerant and SRF-intolerant treat-and-extend regimens for neovascular age-related macular degeneration analyze the association with best-corrected visual acuity.Macular (SRF intraretinal fluid) was quantified on optical coherence tomography volumetric scans using a trained validated deep learning algorithm. Fluid complete resolution automatically assessed throughout study. The impact of location acuity computed...

Abstract Diabetic macular edema (DME) and retina vein occlusion (RVO) are diseases in which central photoreceptors affected due to pathological accumulation of fluid. Optical coherence tomography allows visually assess evaluate photoreceptor integrity, whose alteration has been observed as an important biomarker both diseases. However, the manual quantification this layered structure is challenging, tedious time-consuming. In paper we introduce a deep learning approach for automatically...

Robust forecasting of the future anatomical changes inflicted by an ongoing disease is extremely challenging task that out grasp even for experienced healthcare professionals. Such a capability, however, great importance since it can improve patient management providing information on speed progression already at admission stage, or enrich clinical trials with fast progressors and avoid need control arms means digital twins. In this work, we develop deep learning method models evolution...

To investigate associations between residual subretinal fluid (rSRF) volumes, quantified using artificial intelligence and treatment outcomes in a (SRF)-tolerant treat-and-extend (T&E) regimen neovascular age-related macular degeneration.Patients enrolled the prospective, multicenter FLUID study randomized an SRF-tolerant T&E were examined by spectral-domain optical coherence tomography tested for best-corrected visual acuity (BCVA). Intraretinal SRF volumes tools. In total, 375 visits of 98...

Purpose: To quantify morphologic photoreceptor integrity during anti–vascular endothelial growth factor (anti-VEGF) therapy of neovascular age-related macular degeneration and correlate these findings with disease morphology function. Methods: This presents a post hoc analysis on spectral-domain optical coherence tomography data 185 patients, acquired at baseline, Month 3, 12 in multicenter, prospective trial. Loss the ellipsoid zone (EZ) was manually quantified all volumes. Intraretinal...

Purpose: The subsidence of the outer plexiform layer (OPL) is an important imaging biomarker on optical coherence tomography (OCT) associated with early retinal atrophy and a risk factor for progression to geographic in patients intermediate age-related macular degeneration (AMD). Deep neural networks (DNNs) OCT can support automated detection localization this biomarker. Methods: method predicts potential OPL locations OCTs. A module (DM) infers bounding boxes around subsidences likelihood...

In patients with age-related macular degeneration (AMD), the risk of progression to late stages is highly heterogeneous, and prognostic imaging biomarkers remain unclear. We propose a deep survival model predict towards atrophic stage AMD. The combines advantages modelling, accounting for time-to-event censoring, learning, generating prediction from raw 3D OCT scans, without need extracting predefined set quantitative biomarkers. demonstrate, in an extensive evaluations, based on two large...

The use of orthogonal projections on high-dimensional input and target data in learning frameworks is studied. First, we investigate the relations between two standard objectives dimension reduction, preservation variance pairwise relative distances. Investigations their asymptotic correlation as well numerical experiments show that a projection does usually not satisfy both at once. In classification problem determine balance compare subsequent results. Next, extend our application to deep...

The use of multimodal imaging has led to significant improvements in the diagnosis and treatment many diseases. Similar clinical practice, some works have demonstrated benefits fusion for automatic segmentation classification using deep learning-based methods. However, current methods are limited modalities with same dimensionality (e.g., 3D+3D, 2D+2D), which is not always possible, strategies implemented by incompatible localization tasks. In this work, we propose a novel framework data...

PURPOSE OF THE STUDyPercutaneous plating of the distal tibia via a limited incision is an accepted technique osteosynthesis for extra-articular and simple intra-articular fractures.The aim this study was to analyze structures that are at risk during approach. MATERIAL AnD METHODSThirteen unpaired adult lower limbs were used study.Thirteen, 15-hole LCP anterolateral tibial plates percutaneously inserted according recommended technique.Dissection performed examine relation superficial deep...

Abstract This study aims to evaluate the relation between lumbosacral trunk (LT) and sacro-iliac joint (SIJ). In forty anatomic specimens (hemipelves) a classical antero-lateral approach SIJ was performed. The marked at linea terminalis (reference point A). Reference B situated upper edge of interosseous ligament. length (distance A B) distance ventral branch fourth (L4) fifth (L5) lumbar spinal nerves were measured. had mean 58.0 mm. L5 located closer in very long SIJs (mean length: ≥ 6.5...

Abstract To evaluate the risk of iatrogenic injury when using a dual-incision minimally invasive technique to decompress anterior and peroneal compartments lower leg. Forty extremities from 20 adult cadavers, embalmed with Thiel’s method, were subject fasciotomy compartment fasciotomy. The first incision was made 12 cm proximal lateral malleolus identify protect superficial nerve (SPN). second at mid-point Fibula (half-way between fibular head malleolus). Release successful in all specimens....

The aim of our study was to project the borders flexor retinaculum (FR) onto superficial landmarks since its insufficient splitting is most common reason for persistence symptoms after carpal tunnel release. In 60 hands radial and ulnar styloid processes were linked by a horizontal line longitudinal laid through ring finger's side. These intersected resulting in reference point "A" on forearm. As second basing "B", margin finger at palmar digital crease chosen. Measurement FR carried out...