Latest AI and machine learning research in ophthalmology for healthcare professionals.
We propose VISION-SLS, a method for nonlinear output-feedback control from high-resolution RGB images which provides robust constraint satisfaction guarantees under calibrated uncertainty bounds despite partial observability, sensor noise, and nonlinear dynamics. To enable scalability while retaining guarantees, we propose: (i) a learned low-dimensional observation map from pretrained visual featu...
Diffuse Large B-cell lymphoma (DLBCL) is the most common aggressive lymphoma in the Western world. First-line immunochemotherapy fails in approximately 30-40% of patients, with refractory and relapse patients presenting a dismal prognosis. Currently, these high-risk patients cannot be accurately identified at diagnosis. Using statistical modeling and machine learning approaches applied to large pu...
Real-time assessment of human pluripotent stem cell (hPSC) quality is critical for reproducibility and safety in regenerative medicine, yet current me...
Accurate biomedical image classification under low-resource conditions remains challenging due to limited annotations, subtle inter-class visual diffe...
The Universal Approximation Theorem (UAT) guarantees universal function approximation but does not explain how residual models distribute approximatio...
Vision-language models (VLMs) are increasingly deployed as evaluators in tasks requiring nuanced image understanding, yet their reliability in scoring...
Content-based image retrieval (CBIR) systems enable users to search images based on visual content instead of relying on metadata. The text domain has...
Unified multimodal models typically rely on pretrained vision encoders and use separate visual representations for understanding and generation, creat...
Noisy labels are a pervasive challenge in medical image classification, where annotation errors arise from inter-observer variability and diagnostic a...
Space-biology imaging studies are often constrained by severe data scarcity, limiting the development of robust machine-learning biomarkers. Rodent sp...
Urban transportation systems face growing safety challenges that require scalable intelligence for emerging smart mobility infrastructures. While rece...
Self-supervised learning has achieved remarkable success in learning visual representations from clean data, yet remains challenging when clean observ...
Multimodal learning has the potential to improve clinical prediction by integrating complementary data sources, but the incremental value of imaging b...
Every ML kernel ships with an implicit contract about what it computes. People rarely write the contract down. When two kernels disagree -- when a mat...
Entrenchment - epistasis that locks in amino acid differences between homologous proteins, so each disfavors substitutions toward the other's state - ...
Vision Graph Neural Networks (ViGs) represent an image as a graph of patch tokens, enabling adaptive, feature-driven neighborhoods. Unlike CNNs with f...
Vision--language models (VLMs) often fail on abstract visual reasoning benchmarks such as Bongard problems, raising the question of whether the main b...
The advancement of Large Vision-Language Models (LVLMs) requires precise local region-based reasoning that faithfully grounds the model's logic in act...
Large Vision-Language Models (VLMs) are increasingly used to evaluate outputs of other models, for image-to-text (I2T) tasks such as visual question a...
We address the ambiguities in the super-resolution problem under translation. We demonstrate that combinations of low-resolution images at different s...