PulseAugur / Brief
EN
LIVE 10:43:21

Brief

last 24h
[5/5] 224 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

  1. EyeMVP: OCT-Informed Fundus Representation Learning via Paired CFP--OCT Pretraining

    Researchers have developed EyeMVP, a novel foundation model for retinal image analysis that integrates data from both color fundus photography (CFP) and optical coherence tomography (OCT). Pretrained on a large dataset of paired CFP-OCT images from over 112,000 patients, EyeMVP learns to enhance CFP representations with OCT-derived structural information. This allows for more accurate diagnoses using only CFP images during inference, showing improved performance on tasks like macular edema and myopic macular schisis detection compared to existing models and human ophthalmologists in exploratory studies. AI

  2. Deep Learning-assisted AMD Staging based on OCT and OCT Angiography

    Researchers have developed deep learning models to automatically stage age-related macular degeneration (AMD) using optical coherence tomography (OCT) and OCT angiography (OCTA) data. The models demonstrated strong performance in grading AMD severity, with substantial agreement with a reference standard. A biomarker-based model showed the highest overall performance and was particularly effective at detecting early AMD. AI

    IMPACT Novel deep learning approach could improve early detection and management of age-related macular degeneration.

  3. OphMAE: Bridging Volumetric and Planar Imaging with a Foundation Model for Adaptive Ophthalmological Diagnosis

    Researchers have developed OphMAE, a novel foundation model for ophthalmological diagnosis that integrates both 3D Optical Coherence Tomography (OCT) and 2D en face OCT imaging. Pre-trained on over 183,000 OCT images, OphMAE achieved state-of-the-art performance on 17 diagnostic tasks, including an AUC of 96.9% for Age-related Macular Degeneration (AMD). The model demonstrates adaptability by maintaining high accuracy even with single-modality 2D inputs and showing strong performance with limited labeled data. AI

    OphMAE: Bridging Volumetric and Planar Imaging with a Foundation Model for Adaptive Ophthalmological Diagnosis

    IMPACT This adaptable AI framework could improve diagnostic capabilities in ophthalmology, especially in resource-limited settings.

  4. SAIL: Structure-Aware Interpretable Learning for Anatomy-Aligned Post-hoc Explanations in OCT

    Researchers have developed a new framework called SAIL (Structure-Aware Interpretable Learning) to improve the explainability of deep learning models used in optical coherence tomography (OCT) for retinal disease diagnosis. Existing methods often fail to accurately delineate anatomical structures or respect boundaries, hindering clinical trust. SAIL integrates anatomical priors with semantic features to produce sharper, more clinically meaningful, and anatomy-aligned explanations without altering standard post-hoc explainability techniques. AI

    SAIL: Structure-Aware Interpretable Learning for Anatomy-Aligned Post-hoc Explanations in OCT

    IMPACT Enhances trust and clinical adoption of AI in medical diagnostics by providing more reliable and interpretable explanations.

  5. ‘Off Campus’ Has Set A Rotten Tomatoes Score Record For The Past Year

    The romance genre show "Off Campus" has achieved a record-breaking combined critic and audience score on Rotten Tomatoes for the past year. It narrowly surpassed other popular series like "Heated Rivalry" and "Forever" with a total score of 186. The show, available on Amazon, also ranks as the third highest-rated series overall on the platform for the year, demonstrating significant critical and audience acclaim. AI

    ‘Off Campus’ Has Set A Rotten Tomatoes Score Record For The Past Year