ConferenceProgress in Biomedical Optics and Imaging Proceedings of SPIE · April 2, 2026
Current radiology study harmonization methods rely heavily on textual metadata from DICOM headers, creating significant vulnerabilities when descriptions are incomplete, corrupted, or missing. This research introduces a Multi-Granular Vision-Language Align ...
Full textCite
Conference13th International Conference on Learning Representations Iclr 2025 · January 1, 2025
Illumination degradation image restoration (IDIR) techniques aim to improve the visibility of degraded images and mitigate the adverse effects of deteriorated illumination. Among these algorithms, diffusion-based models (DM) have shown promising performanc ...
Cite
ConferenceProceedings of Machine Learning Research · January 1, 2025
Existing concealed object segmentation (COS) methods frequently utilize reversible strategies, e.g., background reversible attention and foreground-background reversible calibration, to address uncertain regions. However, these approaches are typically res ...
Cite
ConferenceProgress in Biomedical Optics and Imaging Proceedings of SPIE · January 1, 2023
Adaptive optics optical coherence tomography (AO-OCT) has allowed for the reliable 3-D imaging of individual retinal cells. The current AO-OCT systems are limited to tabletop implementation due to their size and complexity. This work describes the design a ...
Full textCite
ConferenceProceedings of the 38th Conference on Uncertainty in Artificial Intelligence Uai 2022 · January 1, 2022
We propose a neural network architecture that enables non-parametric calibration and generation of multivariate extreme value distributions (MEVs). MEVs arise from Extreme Value Theory (EVT) as the necessary class of models when extrapolating a distributio ...
Cite
ConferenceProceedings of Machine Learning Research · January 1, 2022
We propose a neural network architecture that enables non-parametric calibration and generation of multivariate extreme value distributions (MEVs). MEVs arise from Extreme Value Theory (EVT) as the necessary class of models when extrapolating a distributio ...
Cite
ConferenceProceedings of Machine Learning Research · January 1, 2021
It has been conjectured that the Fisher divergence is more robust to model uncertainty than the conventional Kullback-Leibler (KL) divergence. This motivates the design of a new class of robust generative auto-encoders (AE) referred to as Fisher auto-encod ...
Cite
ConferenceOptics Infobase Conference Papers · January 1, 2021
We present 3D optical coherence refraction tomography (OCRT), which incoherently synthesizes an extended 3D transfer function using multi-view OCT volumes spanning two angular dimensions. Our setup features a novel use of parabolic mirrors, achieving milli ...
Cite
ConferenceProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition · January 1, 2021
We present a feature-free photogrammetric technique that enables quantitative 3D mesoscopic (mm-scale height variation) imaging with tens-of-micron accuracy from sequences of images acquired by a smartphone at close range (several cm) under freehand motion ...
Full textCite
ConferenceOphthalmology · June 2020
PurposeTo validate the efficacy of a fully automatic, deep learning-based segmentation algorithm beyond conventional performance metrics by measuring the primary outcome of a clinical trial for macular telangiectasia type 2 (MacTel2).Design
Full textCite
ConferenceOptics Infobase Conference Papers · January 1, 2020
We present a preliminary k-space-based analysis of OCT and our recently proposed extension, optical coherence refraction tomography (OCRT). Our theory provides insight into how angular compounding enhances resolution and reduces speckle in coherence imagin ...
Cite
ConferenceProgress in Biomedical Optics and Imaging - Proceedings of SPIE · January 1, 2020
Quantitative features of individual ganglion cells (GCs) are potential paradigm changing biomarkers for improved diagnosis and treatment monitoring of GC loss in neurodegenerative diseases like glaucoma and Alzheimer's disease. The recent incorporation of ...
Full textCite
ConferenceProgress in Biomedical Optics and Imaging Proceedings of SPIE · January 1, 2020
Adaptive optics scanning laser ophthalmoscopy (AOSLO) has advanced the study of retinal structure and function by enabling in vivo imaging of individual photoreceptors. Most implementations of AOSLOs are large, complex tabletop systems, thereby preventing ...
Full textCite
ConferenceIEEE International Ultrasonics Symposium Ius · October 1, 2019
Image post-processing is used in clinical-grade ultrasound scanners to improve image quality (e.g., reduce speckle noise and enhance contrast). These post-processing techniques vary across manufacturers and are generally kept proprietary, which presents a ...
Full textCite
ConferenceIs and T Archiving Conference Archiving 2018 · January 1, 2018
This rapid development and testing project captured data from multiple digital imaging techniques to try to see texts in papyrus mâché mummy mask cartonnage layers. Prior studies by other scholars destroyed the masks to access the papyri, denying future re ...
Cite
ConferenceOphthalmology · December 2017
PurposeAppearance of geographic atrophy (GA) on color photography (CP) is preceded by specific features on spectral-domain optical coherence tomography (SD OCT). We aimed to build SD OCT-based risk assessment models for 5-year new onset of GA and ...
Full textCite
ConferenceDigital Holography and Three Dimensional Imaging Dh 2012 · December 1, 2012
We report on a preliminary study comparing corneal refractive power acquired using distributed scanning OCT (DSOCT) to corneal topography and Scheimpflug photography in normal subjects both before and after refractive surgery. ...
Cite
