A5089886529- Retinal Imaging and Analysis
- Retinal Diseases and Treatments
- Glaucoma and retinal disorders
- Retinal and Optic Conditions
- Optical Coherence Tomography Applications
- Digital Imaging for Blood Diseases
- Robotics and Sensor-Based Localization
- Advanced Neural Network Applications
- Advanced Electron Microscopy Techniques and Applications
- Electron and X-Ray Spectroscopy Techniques
- Advanced Image and Video Retrieval Techniques
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Ocular Oncology and Treatments
- Retinal and Macular Surgery
- Angiogenesis and VEGF in Cancer
- Integrated Circuits and Semiconductor Failure Analysis
- Biometric Identification and Security
- Optical measurement and interference techniques
- Brain Tumor Detection and Classification
- Coronary Interventions and Diagnostics
- Cell Image Analysis Techniques
- Force Microscopy Techniques and Applications
- Image Processing Techniques and Applications
- Artificial Intelligence in Healthcare
Retina Associates
2025
Auris Medical (Switzerland)
2020-2021
University of Bern
2012-2017
École Polytechnique Fédérale de Lausanne
2017
Institute of Physics
2017
Retinal swelling due to the accumulation of fluid is associated with most vision-threatening retinal diseases. Optical coherence tomography (OCT) current standard care in assessing presence and quantity image-guided treatment management. Deep learning methods have made their impact across medical imaging, many OCT analysis been proposed. However, it currently not clear how successful they are interpreting on OCT, which lack standardized benchmarks. To address this, we organized a challenge...
We introduce a novel CNN-based feature point detector - Greedily Learned Accurate Match Points (GLAMpoints) learned in semi-supervised manner. Our extracts repeatable, stable interest points with dense coverage, specifically designed to maximize the correct matching specific domain, which is contrast conventional techniques that optimize indirect metrics. In this paper, we apply our method on challenging retinal slitlamp images, for classical detectors yield unsatisfactory results due low...
To assess the potential of machine learning to predict low and high treatment demand in real life patients with neovascular age-related macular degeneration (nAMD), retinal vein occlusion (RVO), diabetic edema (DME) treated according a treat-and-extend regimen (TER). Retrospective cohort study. Three hundred seventy-seven eyes (340 patients) nAMD 333 (285 RVO or DME anti–vascular endothelial growth factor agents (VEGF) predefined TER from 2014 through 2018. Eyes were grouped by disease into...
Aim To evaluate the impact of fluid volume fluctuations quantified with artificial intelligence in optical coherence tomography scans during maintenance phase and visual outcomes at 12 24 months a real-world, multicentre, national cohort treatment-naïve neovascular age-related macular degeneration (nAMD) eyes. Methods Demographics, acuity (VA) number injections were collected using Fight Retinal Blindness tool. Intraretinal (IRF), subretinal (SRF), pigment epithelial detachment (PED), total...
To illustrate the treatment effect of Pegcetacoplan for atrophy secondary to age-related macular degeneration (AMD), on an individualized topographic progression prediction basis, using a deep learning model. Patients (N = 99) with AMD longitudinal optical coherence tomography (OCT) data were retrospectively analyzed. We used previously published deep-learning-based algorithm predict 2-year progression, including likelihood future retinal pigment epithelial and outer (RORA), according...
Purpose: To develop a reliable algorithm for the automated identification, localization, and volume measurement of exudative manifestations in neovascular age-related macular degeneration (nAMD), including intraretinal (IRF), subretinal fluid (SRF), pigment epithelium detachment (PED), using deep-learning approach. Methods: One hundred seven spectral domain optical coherence tomography (OCT) cube volumes were extracted from nAMD eyes. Manual annotation IRF, SRF, PED was performed. Ninety-two...
Aim To explore associations between artificial intelligence (AI)-based fluid compartment quantifications and 12 months visual outcomes in OCT images from a real-world, multicentre, national cohort of naïve neovascular age-related macular degeneration (nAMD) treated eyes. Methods Demographics, acuity (VA), drug number injections data were collected using validated web-based tool. Fluid including intraretinal (IRF), subretinal (SRF) pigment epithelial detachment (PED) the fovea (1 mm),...
Abstract In this work we evaluated a postprocessing, customized automatic retinal OCT B-scan enhancement software for noise reduction, contrast and improved depth quality applicable to Heidelberg Engineering Spectralis devices. A trained deep neural network was used process images from an dataset with ground truth biomarker gradings. Performance assessed by the evaluation of two expert graders who image clear preference enhanced over original images. Objective measures such as SNR estimation...
Abstract Age-related macular degeneration (AMD) is a progressive retinal disease, causing vision loss. A more detailed characterization of its atrophic form became possible thanks to the introduction Optical Coherence Tomography (OCT). However, manual atrophy quantification in 3D scans tedious task and prevents taking full advantage accurate retina depiction. In this study we developed fully automated algorithm segmenting Retinal Pigment Epithelial Outer Atrophy (RORA) dry AMD on OCT. 62...
In this retrospective cohort study, we wanted to evaluate the performance and analyze insights of an artificial intelligence (AI) algorithm in detecting retinal fluid spectral-domain OCT volume scans from a large patients with neovascular age-related macular degeneration (AMD) diabetic edema (DME).A total 3,981 volumes 374 AMD 11,501 811 DME were acquired Heidelberg-Spectralis device (Heidelberg Engineering Inc., Heidelberg, Germany) between 2013 2021. Each was annotated for presence or...
Purpose: To develop and validate an automatic retinal pigment epithelial outer atrophy (RORA) progression prediction model for nonexudative age-related macular degeneration (AMD) cases in optical coherence tomography (OCT) scans. Methods: Longitudinal OCT data from 129 eyes/119 patients with RORA was collected separated into training testing groups. automatically segmented all scans additionally manually annotated the test OCT-based features such as layers thicknesses, mean reflectivity, a...
Retinoblastoma and uveal melanoma are fast spreading eye tumors usually diagnosed by using 2D Fundus Image Photography (Fundus) Ultrasound (US). Diagnosis treatment planning of such diseases often require additional complementary imaging to confirm the tumor extend via 3D Magnetic Resonance Imaging (MRI). In this context, having automatic segmentations estimate size distribution pathological tissue would be advantageous towards characterization. Until now, alternative has been manual...
To this day, the slit lamp remains first tool used by an ophthalmologist to examine patient eyes. Imaging of retina poses, however, a variety problems, namely shallow depth focus, reflections from optical system, small field view and non-uniform illumination. For ophthalmologists, use images for documentation analysis purposes, extremely challenging due large image artifacts. reason, we propose automatic retinal video mosaicking, which enlarges reduces amount noise reflections, thus...
Purpose: Diabetic retinopathy (DR) is the leading cause of vision impairment in working-age adults. Automated screening can increase DR detection at early stages relatively low costs. We developed and evaluated a cloud-based tool that uses artificial intelligence (AI), LuxIA algorithm, to detect from single fundus image. Methods: Color images were previously graded by expert readers collected Canarian Health Service (Retisalud) used train LuxIA, deep-learning–based algorithm for more than...
Abstract Purpose To develop a fully automated algorithm for accurate detection of fovea location in atrophic age-related macular degeneration (AMD), based on spectral-domain optical coherence tomography (SD-OCT) scans. Methods Image processing was conducted cohort patients affected by geographic atrophy (GA). SD-OCT images (cube volume) from 55 eyes (51 patients) were extracted and processed with layer segmentation to segment Ganglion Cell Layer (GCL) Inner Plexiform (IPL). Their en face...
To compare drusen volume between Heidelberg Spectral Domain (SD-) and Zeiss Swept-Source (SS) PlexElite Optical Coherence Tomography (OCT) determined by manual automated segmentation methods. Thirty-two eyes of 24 patients with Age-Related Macular Degeneration (AMD) maculopathy were included. In the central 1 3 mm ETDRS circle volumes calculated compared. Drusen was performed using manufacturer algorithms two OCT devices. Then, manually corrected compared finally analyzed customized...
We investigate which spectral domain-optical coherence tomography (SD-OCT) setting is superior when measuring subfoveal choroidal thickness (CT) and compared results to an automated segmentation software.Thirty patients underwent enhanced depth imaging (EDI)-OCT. B-scans were extracted in six different settings (W+N = white background/normal contrast 9; W+H background/maximum 16; B+N black 12; B+H C+N Color-encoded image on background at predefined of 9, C+H high/maximal 16), resulting 180...